Laura Pla Pagà EFFECTS of HESPERIDIN on CARDIOVASCULAR DISEASE RISK BIOMARKERS: an OMICS SCIENCE APPROACH

Home , Arame

Laura Pla Pagà

EFFECTS OF HESPERIDIN ON CARDIOVASCULAR

DISEASE RISK BIOMARKERS: AN OMICS SCIENCE

APPROACH

INTERNATIONAL DOCTORAL THESIS

Supervised by:

PhD Rosa M Valls Zamora

PhD Anna Pedret Figuerola

Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut,

Functional Nutrition, Oxidation and Cardiovascular Disease Group (NFOC-

Salut); and Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i

Salut.

Reus, Tarragona, Spain 2020

FAIG CONSTAR que aquest treball, titulat “Effects of hesperidin on cardiovascular disease risk biomarkers: an omics science approach”, que presenta Laura Pla Pagà per a l’obtenció del títol de Doctor amb menció internacional, ha estat realitzat sota la meva direcció al

Departament de Medicina i Cirurgia d’aquesta universitat.

HAGO CONSTAR que el presente trabajo, titulado “Effects of hesperidin on cardiovascular disease risk biomarkers: an omics science approach”, que presenta Laura Pla Pagà para la obtención del título de

Doctor con mención internacional, ha sido realizado bajo mi dirección en el Departamento de Medicina y Cirugía de esta universidad.

I STATE that the present study, entitled “Effects of hesperidin on cardiovascular disease risk biomarkers: an omics science approach”, presented by Laura Pla Pagà for the award of the degree of Doctor with

3 an international distinction, has been carried out under my supervision at the Department of Medicine and Surgery of this university.

Reus, 14 de novembre de 2020 / Reus, 14 de noviembre de 2020 / Reus,

2020 November 14

Les directores de la tesi doctoral

Las directoras de la tesis doctoral

Doctoral thesis supervisors

Dra Rosa M Valls Zamora Dra Anna Pedret Figuerola

ACKNOWLEDGMENTS

First, I would like to thank my mother Basilisa and my father Josep very much for their unconditional support throughout my academic training. Without them, the present doctoral thesis would never have been possible.

Second, I am very grateful to Rosa M Valls for hosting me five years ago when I wanted to know the world of nutrition and health research. Without you, Rosa M Valls, and Rosa Solà and Anna Pedret, I would never have had the professional career I wanted. Thank you very much. You three are an inspiration for me and an example to follow.

Third, I thank Universitat Rovira i Virgili for granting me an international fellowship that allowed me to realize the international stay in France.

Finally, I would like to thank Eurecat, Centre Tecnològic de Catalunya, for the trust placed in me.

“What we know is a drop, what we don’t know is an ocean”

Isaac Newton

ABBREVIATIONS

AHA/ACA American Heart Association/American College of Cardiology AMPK AMP-activated protein kinase BCAA branched-chain amino acids BMI body mass index BP blood pressure CD control drink COVID-19 Coronavirus Disease 2019 CVDRFs cardiovascular diseases risk factors CVD cardiovascular diseases DBP diastolic blood pressure DNA deoxyribonucleic acid ELISA enzyme-linked immunosorbent assay EOJ hesperidin-enriched orange juice ESC European Cardiology Society FMD flow-mediated dilation GCPICH Good Clinical Practice Guidelines of the International Conference of Harmonization HDL-c high-density lipoprotein-cholesterol HS hesperidin supplementation HOMA homeostasis model assessment IL-6 interleukin-6 IPA ingenuity pathway analysis LDL-c low-density lipoprotein-cholesterol LDL-p low-density lipoprotein particles lncRNA long non-coding ribonucleic acid mRNA messenger ribonucleic acid M-S mass spectrometry

NF-κB nuclear factor-κB NO nitric oxide OJ orange juice OPLS-DA orthogonal projection to latent structures discriminant analysis PBMCs peripheral blood mononuclear cells PGC-1α peroxisome proliferator-activated receptor gamma coactivator 1-alpha PLS-DA partial least squares discriminant analysis PP pulse pressure PPAR-γ peroxisome proliferator-activated receptor gamma PRISMA Preferred Reporting Items for Systematic reviews and Meta-Analysis RCT randomized clinical trial RIPA radioimmunoprecipitation assay buffer RNA ribonucleic acid SBP systolic blood pressure sICAM-1 intercellular adhesion molecule-1 sVCAM-1 soluble vascular cell adhesion molecule-1 TMT tandem mass tag TC total cholesterol TG triglycerides HPLC high-performance liquid chromatography VIP variable importance on projection WHO World Health Organization

INDEX

1. JUSTIFICATION………………………………………...…29

2. INTRODUCTION………………………………………..…35 Chapter 1. A worldwide health problem: cardiovascular disease………………………………………………………...37 1.1. Definition of cardiovascular disease. 1.2. Pathophysiology of atherosclerosis. 1.3. Cardiovascular disease risk factors. 1.4. Prevention of cardiovascular disease. Chapter 2. Diet, nutrition and cardiovascular disease………..45 2.1. Nutrition and cardiovascular health: nutrients. 2.2. Nutrition and cardiovascular health: functional foods and bioactive compounds. Chapter 3. Phenolic compounds…………………………….55 3.1. Classification of phenolic compounds. 3.2. Absorption and metabolism of flavonoids. 3.3. Health effects of flavonoids. 3.4. Mechanisms of action. Chapter 4. The flavonoid hesperidin………………………..64 4.1. Health effects of the flavonoid hesperidin. Chapter 5. Omics science: the future of research………...…73 5.1. Transcriptomics approach. 5.2. Proteomics approach. 5.3. Metabolomics approach.

3. HYPOTHESIS AND OBJECTIVES……………………...81

4. METHODS AND RESULTS………………………………85

Project 1: Effects of hesperidin consumption on cardiovascular risk biomarkers: a systematic review of animal studies and human randomized clinical trials…………………………..…87 Projects 2, 3 and 4: Summarizing the CITRUS randomized controlled trial……………………………………………….129 Project 2: Effects of hesperidin in orange juice on blood and pulse pressure in mildly hypertensive individuals: a randomized clinical trial (CITRUS study)…………………………….….137 Project 3: Effect of the consumption of hesperidin in orange juice on the transcriptomic profile of subjects with pre- and stage 1 hypertension: a randomized controlled trial (CITRUS study) (Editor submitted)…………………………………………..161 Project 4: The consumption of hesperidin in orange juice modulates the plasma and urine metabolic profiles in pre- and stage 1 hypertensive subjects to promote beneficial effects on the cardiovascular system: targeted and nontargeted metabolomics approaches (CITRUS study) (Editor submitted)………..…..221 Project 5: Proteomic analysis of heart and kidney tissues in healthy and metabolic syndrome rats after hesperidin supplementation……………………………………………..266

5. SUMMARY OF RESULTS…………………………….279

6. DISCUSSION………………………………….…...... 295

7. PERSPECTIVES……………………………...... 309

8. CONCLUSIONS……………………………...... 319

9. REFERENCES………………………………………....325

ANNEX

ABSTRACT

INTRODUCTION Currently, cardiovascular diseases (CVDs) are still the main cause of mortality worldwide. Thus, there is a necessity to search for new strategies for the prevention and treatment of CVDs, and the flavonoid hesperidin could have cardioprotective effects. However, the literature is scarce, and some results are controversial. Moreover, the mechanisms of action by which hesperidin exerts its protective effects have not been fully defined, and omics science can study the impact of hesperidin consumption on metabolic pathways to determine how hesperidin exerts its health beneficial effects.

OBJECTIVES The main objective is to evaluate the effects of the consumption of hesperidin in orange juice (OJ) and hesperidin-enriched orange juice (EOJ) on cardiovascular disease risk biomarkers, particularly on blood pressure (BP), and to elucidate the possible mechanisms of action through the application of different omics approaches in human samples: transcriptomics and metabolomics, and proteomics approach in experimental rat models to know its effects on key cardiovascular organs such as heart and kidney.

METHODS AND RESULTS To achieve it, five projects were carried out: 1) a systematic review to known the scientific evidence available on the subject in humans and experimental rat models with cardiovascular disease risk factors (CVDRFs) following the PRISMA 2015 guidelines and PICOS criteria; 2) a randomized, parallel, double-blind, placebo-controlled clinical trial with 159 subjects with pre- and stage 1 hypertension who consumed

500 mL/day for 12 weeks of OJ (392 mg/day of hesperidin), EOJ (670 mg/day of hesperidin) or control drink, performing also two single dose studies of 6 hours, to evaluated its effects on BP (CITRUS study); 3) a transcriptomic analysis realised on peripheral blood mononuclear cells (PBMCs) by the Agilent’ Microarray Platform in a subsample of 37 subjects from the CITRUS study after 12 weeks and also after the single dose studies; 4) metabolomics analysis that included targeted approach performed by high-performance liquid chromatography in plasma and urine samples in a subsample of 129 subjects of the CITRUS study, and nontargeted metabolomic approach performed by nuclear magnetic resonance spectroscopy in serum samples in a subsample of 52 subjects and in urine samples in a subsample of 129 subjects of the CITRUS study; 5) an experimental study with metabolic syndrome rats that consume 100 mg/kg body weight/day of hesperidin supplementation or not hesperidin supplementation was performed to realize a proteomics approach of heart and kidney tissues. The results from the systematic review showed that hesperidin improve lipid profile and blood glucose levels in animal models with CVDRFs. However, no definitive conclusion can be drawn in humans. From CITRUS study, the results indicated that the consumption of hesperidin in OJ and EOJ has beneficial effects by reducing the levels of systolic BP and pulse pressures (PP) in a dose-dependent way in humans. Moreover, transcriptomic approach showed the ability of hesperidin to downregulate pro-inflammatory human genes; metabolomics approach showed the ability of hesperidin to decreases human metabolites related with BP, oxidative stress and inflammation; and finally, proteomics approach showed the ability of hesperidin to changes protein expression

26 related with an improvement of cardiovascular system in rat heart and kidney tissues.

CONCLUSIONS Hesperidin reduces human BP in a dose-dependent way. Thus, the hesperidin enrichment achieved with EOJ, can be a useful co-adjuvant tool for BP and PP management in pre- and stage 1 hypertensive subjects. The mechanisms of action by which hesperidin exerts its beneficial effects can be explained through transcriptomics and metabolomics approaches in humans which demonstrated cardioprotective actions through decreases in pro-inflammatory genes and decreases in serum endogenous metabolites related to BP, oxidative stress and inflammation. Moreover, the proteomic approach realized in kidney and heart tissues of metabolic syndrome rats, showed that hesperidin changes proteomic profiles exerting positive effects on two main organs involved on BP regulation and cardiovascular system, reducing free radical scavenging and lipid and glucose metabolism.

Founding sources: The present work was supported by Eurecat, Centre Tecnològic de Catalunya. This work was also supported by an international fellowship (AEE2017), Universitat Rovira i Virgili, Reus, Spain.

1. JUSTIFICATION

Currently, cardiovascular diseases (CVD) are the main cause of mortality worldwide.(1) Triggering these diseases is tobacco smoking, physical inactivity, and unhealthy diet. Thus, the main risk factors implicated in CVD development are modifiable, meaning that we can act on them and reduce the risk of developing it.(2) However, every year more than 17 million people die worldwide from CVD.(3) There is an ongoing development of new drugs to treat CVD and their intermediate biological factors such as hypertension, hypercholesterolemia, diabetes and obesity, but the partial efficacy, partial intolerance and several adverse effects create the necessity to search for new strategies or coadjutant strategies for the prevention and treatment of CVD.(4) Therefore, the scientific world focuses their research on finding new bioactive compounds from food and beverages with beneficial effects on cardiovascular health. Moreover, the study of their impact on metabolic pathways through the omics science approach to determine their mechanisms of action and to identify new biomarkers of diseases or treatment response is also considered. Omics science refers to a field of study in biological sciences that includes transcriptomics, proteomics, metabolomics or genomics, with the objective of collecting a large number of biological molecules involved in the function of an organism at a defined moment and under certain conditions.(5)

It is well recognized that healthy dietary patterns have significant benefits for CVD treatment and CVD prevention.(6) In this sense, there is the Mediterranean diet, traditionally characterized by a high intake of seasonable vegetables, legumes, fruits, grains, fish and seafood, poultry protein, olive oil, and nuts and low intake of dairy products, red meat, processed meat and sugary drinks. The Mediterranean diet was linked

31 with the improvement in lipid profile, insulin sensitivity and endothelial function in subjects with cardiovascular disease risk factors (CVDRFs), acting on inflammation and antithrombotic and oxidative stress markers.(7) Another example is the DASH diet (Dietary Approaches to Stop Hypertension) characterized by a low consumption of salt and high intake of fruits, vegetables, whole grains, low-fat dairy and lean protein, which has demonstrated significant blood pressure (BP) reductions in subjects with hypertension,(8) which is one of the most important CVDRFs.

In this sense, most of these observed effects on cardiovascular health are attributable to bioactive compounds, and one of the most important are phenolic compounds, founded in plants products and mainly present in species, fruits, vegetables, olive oil, nuts and beverages.(9) Phenolic compounds are classified into two large families: flavonoids and nonflavonoids. Phenolic compounds have gained much interest in this field because of their demonstrated capacity to exert beneficial effects on various diseases, including CVDs,(10) and they have been reported in several randomized controlled trials (RCT), the gold standard studies for efficacy evaluation in clinical research.(11) One of the most studied phenolic compounds related to the Mediterranean diet is phenolic compounds from olive oil. In this sense, it is reported that the consumption of a functional virgin olive oil enriched with olive oil and thyme phenolic compounds enhanced the expression of cholesterol efflux regulators, promoting major high-density lipoprotein function and having a beneficial effect on hypercholesterolaemic subjects.(12) Thus, another randomized controlled clinical trial also demonstrated the cardioprotective effect of virgin olive oil and virgin olive oil enriched

32 with phenolic compounds on oxidation, blood coagulation, lipid transport and immune response.(13) On the other hand, there are some promising phenolic compounds not largely studied until today, and one of them is the flavonoid hesperidin. Hesperidin, naturally present in citrus fruits and the main flavonoid of orange and orange juice (OJ), has been shown in preclinical and clinical studies to have a therapeutic effect in several diseases, including CVDs, due to its anti-inflammatory and antioxidant effects, and its capacity to decrease lipids and improve insulin sensitivity.(14) Moreover, hesperidin could exert promising hypotensive activities.(15,16) However, the literature about hesperidin effects is scarce, and some results are unclear and controversial. Besides, OJ is one of the most consumed beverages throughout the world(17) and is totally integrated into our dietary pattern. However, currently the health impact of fruit juices consumption is controversial.(18) Thus, assessing the beneficial effects of OJ on health can be interesting and relevant. In addition, to the best of our knowledge, the mechanism of action by which hesperidin exerts its possible beneficial effects has not been clearly and fully defined. Finally, no omics approach has been performed to study the effects of hesperidin on the transcriptome, metabolome, or proteome profile, since omics science can shed light on the mechanisms of action.

Therefore, looking at the possible potential of the flavonoid hesperidin on cardiovascular health, the possibility of using hesperidin or hesperidin-rich foods and beverages as a coadjutant treatment or prevention strategy for CVDs, and particularly on BP, and due to the scarcity of the available literature, further research is needed.

Accordingly, a randomized controlled trial will be developed to know hesperidin effects on BP and the most efficient dose of hesperidin in humans. Moreover, to know the mechanisms of action and the biological effects of hesperidin, an omics science approach will be realized in human samples and in organs of experimental rat models.

Figure 1. Graphical representation of the justification section of the present doctoral thesis. Abbreviations: BP, blood pressure; CVDRFs, cardiovascular disease risk factors (own source).

2. INTRODUCTION

CHAPTER 1. A worldwide health problem: cardiovascular disease.

The life expectancy of the world population has been increasing over the decades and currently, the average number of years of life in the world is 72(19), and in Spain it is 83.(20) The increase is due to the eradication of several diseases, the mortality reduction of usual pathologies and changes in daily life such as better food control and hygiene measures.

In this sense, CVDs are one of the most advanced fields in research and contribute to an increase in life expectancy. A few decades ago, heart attack and stroke involved fatal outcomes, but today, the prognosis has changed, and patients can recover and live normally.(21) However, CVDs continue to be the main cause of mortality in the world according to the World Health Organization (WHO) (1) and the main cause of mortality in Spain according to the Instituto Nacional de Estadística.(22) Thus, more people die every year from CVDs than from any other cause. The statistics show that in 2030 more than 23 million people will die from CVDs.(23) Several factors influence the development of CVDs, but the experts mainly highlight the progressive ageing of the population and the loss of healthy lifestyle habits. Age is a risk factor itself, and the possibility of developing CVDs increases with advancing age.(21) However, a change towards a healthier and more active life opens a window of possibilities to transform the course of CVDs.

1. 1 Definition of cardiovascular disease.

CVD is a multifactorial disease consisting of a set of disorders of heart and blood vessels that include:(23) • Hypertension. • Coronary heart disease or heart attack. • Cerebrovascular disease or stroke. • Peripheral vascular disease. • Heart failure. • Rheumatic heart disease. • Congenital heart disease. • Cardiomyopathies.

1. 2 Pathophysiology of atherosclerosis.

Atherosclerosis, an inflammatory disease characterized by oxidative stress and systemic inflammation, is the main cause of the major incidence and mortality of CVD mentioned above.(24) It is known that the formation of atherosclerotic lesions occurs due to local inflammation in the vascular wall that is induced by dyslipidaemia, normally because of high low-density lipoprotein-cholesterol (LDL-c) levels,(25) and is produced because of lipid and leucocyte accumulation in blood vessels producing plaque formation.(26) Gradually, the plaque hardens and causes narrowing of the arteries, restricting blood flow. Later, the plaque can rupture and form a thrombus (blood clot) causing a further blockage of blood flow and therefore, the availability of oxygen to the body’s organs.(26)

Thus, atherosclerosis is characterized by endothelial dysfunction and the accumulation of low-density lipoprotein particles (LDLp), immune cells and necrotic debris in the endothelial space, and endothelial activation triggers the expression of leukocyte adhesion molecules such as E-selectin and P-selectin, glycoproteins such as intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion protein 1 (VCAM-1), chemokines such as monocyte chemoattractant protein-1 (MCP-1), etc.(27) Then, the accumulation of LDLp promotes the release of macrophage colony stimulating factor (M-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF), facilitating monocytes maturation.

Additionally, LDLp turn into oxidized LDL (oxLDL) particles, and they can be recognized by macrophage receptors (CD36 or LOX-1). Consequently, the receptors activate nuclear factor kB (NF-kB) signaling, producing pro-inflammatory cytokines such as interleukin 1 beta and tumour necrosis factor (TNFα), generating foam cells. Foam cells capture oxLDL and lysosomal acid lipase converts esters of cholesterol into free cholesterol and fatty acids. Therefore, M1 macrophages, Th1 cells and some B cell subtypes produce pro- inflammatory cytokines and chemokines promoting atherosclerosis. On the other hand, Bregs, Tregs and M2 macrophages suppress inflammation reducing the size of the plaque and stabilizing the atherosclerotic lesion.(27) The graphical representation of atherosclerosis pathogenesis is shown in Figure 2.

Figure 2. Pathogenesis of atherosclerosis. Source: (Herrero- Fernandez B et al. 2019).(27)

1. 3 Cardiovascular disease risk factors.

A risk factor is defined as the measurable characteristic that is causally associated with an increased rate of a disease; and that is an independent and significant predictor of the risk of presenting a disease.(28) Thus, CVDRFs are those that increase the possibility of developing CVDs.

In this sense, it is known that CVDRFs can be divided into two categories: non-modifiable risk factors and modifiable risk factors.

Non-modifiable risk factors are those that we cannot influence. Factors included in this category are the following:(2) - Family history of CVDs. - Age. - Sex: the male sex has a higher risk of developing CVDs. - Ethnicity: CVD mortality rates are especially high among black men. - Socioeconomic status.

On the other hand, modifiable risk factors are those related to habits or behavior performed by humans, which are susceptible to being modified. In this category, the following factors are included:(2) - Tobacco use. - Physical inactivity. - Physiological stress. - Diet.

Smoking is one of the most preventable risk factors that contributes the most to the development of CVDs.(29) There are group therapies and medication approaches that can be used to stop smoking cigarettes, thus reducing the CVD risk. On the other hand, physical inactivity increases CVD risk by 1.5 times.(30) Regular physical activity protects against CVDs since it reduces body weight, lipid levels, blood glucose and BP levels. Moreover, reducing the progression of atherosclerosis decreases oxidative stress, increases sensitivity to insulin and decreases the incidence of coronary disease. Therefore, it reduces the total and CVD morbidity and mortality.(27)

Finally, the role of diet is crucial in the development and prevention of CVDs since the association between eating habits and CVDs is well known.(30) Unhealthy behavior increase the main risk factors that lead to CVDs such as BP, blood glucose, blood lipids, overweight, obesity and diabetes, resulting in hypertension, hypercholesterolaemia, hypertriglyceridaemia, or metabolic syndrome. Therefore, changes in diet are one of the keys to follow to impact CVDRFs.(31)

In this sense, in Table 1 there is a general description of risk goals and target levels for the most important CVDRFs according to the European Cardiology Society (ESC). Thus, achieving these target levels can prevent the development of CVDs.

Table 1. Risk goals and target levels for the most important cardiovascular diseases risk factors according to the ESC.

Smoking No exposure to tobacco.

Diet Healthy diet low in saturated fat with a focus on wholegrain products, fruits, vegetables, and fish.

Physical activity 3.5-7 hours moderately vigorous physical activity every week or 30 – 60 min most days.

Body weight BMI 20-25 kg/m2. Waist circumference <94 cm in men and <80 cm in women.

BP <140/90 mm Hg.

Lipids -Very high-risk: <70 mg/dL, or a reduction of at least LDL-c 50% if the baseline is between 70 and 135 mg/dL. -High-risk: <70 mg/dL, or a reduction of at least 50% if the baseline is between 100 and 200 mg/dL. -Moderate risk: <100 mg/dL.

-Low risk: <116 mg/dL.

Non-HDL-c -For very high risk: <85 mg/dL. -For high risk: 100 mg/dL. -Moderate risk: 130 mg/dL.

TG <150 mg/dL.

Diabetes HbA1c <7%. (<53 mmol/mol).

Abbreviations: BMI, body mass index; BP, blood pressure; LDL-c, low- density lipoprotein-cholesterol; HDL-c, high density lipoprotein cholesterol; TG, triglycerides; HbA1c, glycated hemoglobin. Adapted from: (Mach F, et al. 2020).(32)

1. 4 Prevention of cardiovascular diseases.

To prevent CVDs, a coordinated pool of activities at the individual or population level to eliminate or reduce CVD incidence to the lowest level is needed.(33) From 1994 to the present, the reference guidelines on CVD prevention are the guidelines of the ESC. The guidelines have been revised several times and the latest version was published in 2020.

These guidelines showed the role of lifestyle including diet in the prevention and treatment of CVDs since the probability of developing

CVDs is associated with unhealthy dietary patterns such as excessive intake of sodium, processed food, added sugars or unhealthy fats, and low intake of vegetables, fruits, and whole grains, among others.(34)

CHAPTER 2. Diet, nutrition and cardiovascular disease.

In this section we will explain the general perspective of the influence of dietary patterns such as the Western diet and Mediterranean diet, nutrients such as sodium and fibre, bioactive compounds such as flavonoids, and functional food such as citrus fruits, on CVDs and their main risk factors.

Dietary habits influence the risk of developing CVDs by influencing CVDRFs, such as blood cholesterol levels, body weight and blood glucose levels.(35) Vast scientific evidence has shown that nutrition might be the most significant preventive factor of CVD death and can reverse heart disease. Additionally, diet can be used to manage excess weight, hypertension, diabetes and dyslipidaemia.(36)

Thus, Table 2 summarizes the characteristics of a healthy diet according to the last update of the European Cardiology Society (ESC) for CVD prevention in clinical practice. For that reason, as dietary-associated risk is the most important behavior factor influencing global health, there is increased experimentation with the use of “food is medicine” interventions to prevent, manage and treat chronic diseases such as CVDs. However, CVDs remain the leading cause of death and disability in developed countries, and the challenge through dietary interventions in CVDs is to create more effective strategies to motivate populations to change their diet and maintain it over time.

Table 2. Characteristics of healthy dietary habits. Saturated fatty acids to account for <10% of total energy intake, through replacement by polyunsaturated fatty acids. Trans unsaturated fatty acids: as little as possible preferably no intake from processed food, and <1% of total energy intake from natural origin. <5 g of salt per day. 30–45 g of fiber per day, preferably from wholegrain products. ≥200 g of fruit per day (2–3 servings). ≥200 g of vegetables per day (2–3 servings). Fish 1–2 times per week, one of which to be oily fish. 30 g unsalted nuts per day. Consumption of alcoholic beverages should be limited to 2 glasses per day (20 g/day of alcohol) for men and 1 glass per day (10 g/day of alcohol) for women. Sugar-sweetened soft drinks and alcoholic beverages consumption must be discouraged.

Adapted from: (Piepoli MF, et al. 2016). (34)

From the 1990s there has been a transition from a nutrient-based to a dietary-based approach for addressing nutritional interventions in CVDs because of the new evidence that emerged from randomized clinical trials and meta-analyses. There is evidence (with few exceptions such as sodium or trans-saturated fatty acids) that single nutrients have effects of limited magnitude on chronic diseases such as

CVDs compared with whole foods or with complex integrated dietary interventions expressed as dietary patterns.(31)

It is known that a westernized diet, characterized by a high intake of proteins derived from processed and read meats, saturated fats, refined grains, sugar, salt and alcohol and a low intake of vegetables and fruits, is associated with an increased risk of metabolic and chronic diseases such as CVDs.(37) On the other hand, the most well-studied dietary patterns that have beneficial effects on health, especially on cardiovascular health and hypertension, are the Mediterranean and Dietary Approach to stop Hypertension (DASH) diets. The composition of the Mediterranean diet and DASH diet are detailed in Table 3.

Both dietary patterns are associated with a lower risk of clinical cardiovascular events.(31,38) However, the first diet that demonstrated a reduced mortality risk for CVDs was the Mediterranean diet. It was realized through a clinical trial with nutritional intervention called the PREDIMED study.(39)

Table 3. The composition of Mediterranean and DASH diets. Mediterranean diet DASH diet

-High content in fresh fruits, -High in vegetables, fruits, low-fat vegetables, whole grains, and fermented dairy products, whole fatty fish (rich in ꞷ-3 PUFA). grains, poultry, fish, and nuts. -Low content in red meat. -Low in sweets, sugar-sweetened -Substituted lower-fat or fat-free beverages, and red meats. dairy products for higher-fat -Low in saturated fat, total fat, and dairy foods. cholesterol. -Olive oil, nuts, or margarines. -Rich in potassium, magnesium, and calcium. -Rich in protein and fiber.

Abbreviations: DASH, Dietary for Approach to Stop Hypertension; PUFA, poly-unsaturated fatty acids. Adapted from: (Ravera A et al. 2016).(40)

The Mediterranean diet, characterized as rich in whole grains, fruit, and vegetables and low in meat, with a considerable amount of fat from olive oil and nuts,(40) was first described in 1979 by the American biologist Ancel Keys when he observed that the population of the shores of the Mediterranean Sea (in Greece, south of Italy and Yugoslavia) had a lower incidence of CVDs. Later, it would be reported that a Mediterranean diet protects against coronary heart diseases and reduces the risk of the development of diabetes mellitus type 2 and metabolic syndrome.(41,42) This type of diet can reduce CVD risk by mechanisms that reduce BP, lipids, glucose, endothelial function, waist circumference and body mass index (BMI).(43) Additionally, it has

48 been reported that a Mediterranean diet can increase nitric oxide (NO) bioavailability and can have antioxidant and anti-inflammatory properties. Moreover, the Mediterranean diet seems to have beneficial effects on the synergy among various cardioprotective nutrients and foods.(43)

On the other hand, in the 1990s, a research group led by Lawrence Appel evaluated the effects of a diet rich in fruit, vegetables and low- fat dairy foods on BP levels in a randomized study called the Dietary Approach to Stop Hypertension (DASH) trial. The DASH diet has demonstrated efficacy in treating hypertension without antihypertensive medication. Moreover, the DASH diet decreases the risk of developing diabetes in randomized controlled clinical trials and reduces cardiovascular mortality in prospective cohort studies.(44)

Both the Mediterranean and DASH diets demonstrated an improvement in large CVDRFs, including long-term weight gain, BP levels, glucose- insulin homeostasis, lipid profile, inflammation and endothelial function.(43,44)

Last, vegetarian diets also showed beneficial effects on cardiovascular health. It is known that vegetarian patterns reduce CVD mortality and the risk of coronary heart disease.(45) In addition, in 2014 a non- controlled study showed that heart disease could be reversed by a low- fat vegan diet.(46) Moreover, recently published review concluded that a vegetarian diet exerts beneficial effects on BP levels, lipid profiles, platelet aggregation, obesity, metabolic syndrome and type 2 diabetes mellitus.(45) All its benefits are the result of the lower exposure to

49 harmful substances contained in animal products such as saturated fats and cholesterol, and greater consumption of whole plants rich in fibre and bioactive compounds.(45)

2. 1 Nutrition and cardiovascular health: nutrients.

The evidence shows that some beneficial effects on several CVDRFs are the higher intake of specific nutrients present in foods and the lower intake of another nutrients.

For example, the intake of long-chain ꞷ-3 polyunsaturated fatty acids and their consequent higher circulant levels are protective against CVDs.(47) This is due to the anti-inflammatory and anti-atherosclerotic effects achieved by altering the fatty acid composition of inflammatory cells.(48) On the other hand, the reduced consumption of saturated fats and trans-fats also shows positive effects on cardiovascular health decreasing LDL-c levels.(49) For that reason, the American Heart Association/American College of Cardiology (AHA/ACA) published in guidelines for the management of reducing CVD risk that included the recommendation of ensuring the intake of polyunsaturated fatty acids and reducing the intake of saturated and trans-fat fatty acids.(50)

On the other hand, the AHA/ACA also reported that the consumption of salt and sodium present naturally in foods and added during cooking can affect the CVDs risk. In this sense, reduced sodium intake is related to lower BP levels in adults, which is useful in preventing or treating hypertension and decreasing the risk of developing CVDs.(50) It is known that a higher dietary sodium intake increases extracellular

50 volume and cardiac output, increasing BP levels. Moreover, combined with sodium intake, abnormal serum levels of potassium and magnesium may affect CVDs because of their influence on hypertension, arrhythmia, and myocardial infarction. This relation is one of the most studied, and in Figure 3, the metabolism of sodium, potassium and magnesium in the body and their relationship with BP levels are presented.

Figure 3. The metabolism of sodium, potassium and magnesium in the body and their relationship with BP levels. Adapted from: (Mohammadifard N et al. 2018).(51)

Moreover, several case-controls and prospective observational studies showed inverse associations between micronutrients such as the levels of vitamin A and vitamin E and the risk of CVDs.(34) Additionally, a correct intake of dietary calcium has beneficial effect on the cardiovascular system through the improvement in lipid profile, BP, obesity, insulin secretion, inflammation and anti-thrombotic agents.(51) It is known that micronutrients can exert their protective effect in three forms: decreasing endothelial cell damage, increasing NO production and/or inhibiting the oxidation of LDL-p. (24)

Additionally, prospective cohorts and meta-analyses observe that a higher intake of total fibre decrease the risk of coronary artery disease, stroke and diabetes mellitus type 2, decreasing blood cholesterol levels and BP, while a deficiency of fiber intake was associated with the development of CVDs.(24)

There is much scientific evidence that relates the beneficial effects of certain nutrients to CVDs and their risk factors. Therefore, the recognition of appropriate nutrients and their adequate intake have an important role in preventing CVDs. It is important to focus on the possible beneficial effects of specific nutrient intake to avoid possible deficiencies in nutrients that can lead to the development of atherosclerotic disease.

2. 2 Nutrition and cardiovascular health: bioactive compounds and functional foods.

The term “functional food” was coined in Japan and the USA in 1970.(52) Functional foods are foods fortified with different probiotics and microorganisms, or natural or processed foods that have biologically active compounds and specific health-promoting benefits that have been scientifically substantiated.(52)

Table 4 summarizes some of the most studied functional foods, their respective bioactive compounds, and their potential mechanism of action.

Table 4. Bioactive compounds, functional foods, and potential mechanisms. Bioactive compound: Functional food: Potential mechanism: Tocopherols, Nuts Lowering blood ꞷ-3 fatty acids cholesterol Fiber Legumes Inhibition of LDL-p oxidation Genistein and daidzein Soy proteins Antioxidant action and platelet aggregation Flavonoids Dark chocolate Lowering blood TG, decreasing BP and antioxidant action Lycopene Tomato Antioxidant action Ascorbic acid Citrus Antioxidant action

Anthocyanins and Grapes Endothelial function, catechins antioxidant action and platelet aggregation Phytochemicals Whole grains Lowering blood TG and decreasing BP Abbreviations: LDL-p, low-density lipoprotein particles; TG, triglycerides; BP, blood pressure. Adapted from: (Asgary S et al. 2018).(52)

The presence of some bioactive compounds present naturally in food and beverages is also responsible for the prevention and treatment of CVDs because of their beneficial effects on atherosclerosis development, reducing LDL-c levels, inflammation, and oxidative stress.(9)

In recent decades, the number of studies evaluating the physiological activities of bioactive compounds from food has markedly increased, and phenolic compounds are one of the most studied. However, due to the large types of phenolic compounds and controversial results reported, more research is needed in this area since the identification of beneficial effects on cardiovascular health through phenolic compound intake could be part of the new nutritional treatment tools.

CHAPTER 3. Phenolic compounds.

In this section, the definition and summary of the characteristics and effects of phenolic compounds will be reported, focusing on flavonoids since they are studied in the present doctoral thesis.

Phenolic compounds are secondary metabolites of plants present in high concentrations in several species, food, and beverages. In contrast to minerals, vitamins and other nutrients, the deficiency of phenolic compounds intake does not produce specific deficiency diseases, although adequate phenolic compounds intake has beneficial effects on health status, especially on chronic diseases.(53)

Currently, phenolic compounds are one of the most studied bioactive compounds because they are the most consumed antioxidant.(54) Total phenolic compounds ingestion, according to our previous results, ranged from 1196.7 to 1967.9 mg/day in an adult healthy Mediterranean population.(55) It is known that cocoa, tea, fruits, and vegetables are the main foods rich in phenolic compounds (56), and a high intake of specific phenolic compounds has been linked to reduced mortality from specific vascular diseases and cancers.(57)

3. 1 Classification of phenolic compounds.

There are more than 8000 different types of phenolic compounds that differ between them depending on the number of phenolic rings they have, and the substituent attached to the rings.(9) Phenolic compounds are classified into two large categories: flavonoids and nonflavonoids.

Flavonoids: Flavonoids are the most common type in plants and consist of 15 carbons with 2 aromatic rings connected by a 3-carbon bridge (Figure 4).

The main subclasses of flavonoids are flavonols, flavones, flavan-3-ols, isoflavones, flavanones and anthocyanidins. Yellow and red onions are especially rich sources of flavonols containing high concentrations of quercetin; rooibos tea and caffeine-free beverages have high concentrations of flavones such as apigenin and luteolin; leguminous plants are especially rich sources of isoflavones containing high concentrations of daidzein and genistein; citrus fruits have high concentrations of flavanones such as naringenin and hesperetin; and colourful fruits have high concentrations of the anthocyanidins cyanidin and pelargonidin.(9)

Non-flavonoids: The main non-flavonoids present in food are phenolic acids, and gallic acid is the most common. Non-flavonoids also include stilbenes and lignans (Figure 5).

The main stilbenes are resveratrol, present in red wine and blueberries, while secoisolariciresinol is the most common lignan, present in linseed, cereals, and grain.(9)

Figure 4. Structure of flavonoid and their subclasses. Source: (Del Rio D et al. 2013).(9)

Figure 5. Structure of the main non-flavonoids: gallic acid (phenolic acid), resveratrol (stilbene) and enterodiol (lignan). Adapted from: (Del Rio D et al. 2013).(9)

3. 2 Absorption and metabolism of flavonoids.

Flavonoids present in the diet are mainly present in their glycoside form, and deglycosylation takes place in the small and large intestines, depending on the type of sugar moiety. Two enzymes have been reported to act as β-glucosidases in the small intestine against flavonoid monoglucosides: lactase-phlorizin hydrolase (LPH) and cytosolic β- glucosidase.(9)

An example of the absorption and metabolism of flavonoids, specifically apple flavonoids, is represented in Figure 6.

Figure 6. Absorption and metabolism of flavonoids. Quercetin glycosides represent the absorption and metabolism of flavonoid glycosides. Epicatechin represents the absorption and metabolism of the flavonoid subclass, the flavan-3-ols. Abbreviations: SGLT, sodium- dependent glucose transporter; LPH, lactase-phlohizin-hydrolase; CBG, cytosolic β-glucosidase; MRP, multidrug resistance protein; P- gp, P-glycoprotein. Source: (Bondonno CP et al. 2015).(58)

3. 3 Health effects of flavonoids.

At the end of the 19th century the chemical structure of flavonoids was identified and in the early years of the 20th century, flavonoids and their related substances were synthesized in the laboratory. Finally, in the 1930s Albert Szent-Györgyi focused his attention on the effects of specific flavonoids on human health.(59)

Since the effects of flavonoid consumption have been studied for decades, in vivo and in vitro studies were performed to determine their mechanisms of action and their implications on biological pathways. Additionally, epidemiological studies and clinical trials in humans were performed to evaluate the effects of flavonoid consumption.

Epidemiological evidence has demonstrated a reduction in the risk of fatal CVDs in subjects with a high flavonoid intake. Moreover, prospective studies observed an association between high flavonoid intake and a lower risk of CVD mortality and future cardiovascular events.(60) On the other hand, human randomized clinical trials showed protective effects of flavonoid consumption against CVDs. For example, flavonoids of phenol-enriched olive oils modulate oxidative balance producing cardioprotective effects,(61) and flavonoids of flavonoid- rich apple improve endothelial function decreasing the risk for CVDs.(62) In this sense, several studies have shown benefits of flavonoid consumption on the prevention and treatment of several lifestyle-related diseases,(56) including atherosclerosis,(63) coronary heart diseases(64) and metabolic syndrome.(65)

Meta-analyses reported that the consumption of flavonoid-rich foods is associated with a reduced risk of cause-specific mortalities such as those attributable to cancer, diabetes, and CVDs.(57) Additionally, observational studies showed a lower risk of cardiometabolic events with a high dietary intake of flavonoids.(66,67) In human clinical trials, flavonoid consumption also shows beneficial effects on several parameters related to chronic diseases. For example, RCT performed on subjects with CVDRFs such as hypertension, overweight or obesity, diabetes and dyslipidaemia, reported that the daily consumption of chocolate containing 70% cocoa, rich in gallic acid, for 6 weeks improved biochemical parameters (such as total cholesterol (TC), LDL- c and triglycerides (TG)) and waist circumference.(68) In another randomized clinical trial, both the single dose and chronic consumption of flavonoid-rich apple improved the endothelial function, an independent risk factor for CVDs.(62) Moreover, after daily consumption for 8 weeks of blueberry, rich in anthocyanins, the systolic BP levels were reduced in postmenopausal women with pre and stage 1 hypertension, due to the increase in NO production.(69) A recently published review shows that flavonoid-enriched foods decrease BP levels and improve endothelial function by promoting vascular dilation in geriatric patients, who are characterized by multiple chronic diseases.(70) Furthermore, consumption of OJ, which is naturally rich in hesperidin, for 4 weeks, increases endothelium- dependent microvascular reactivity and decreases diastolic BP.(71)

Thus, several clinical trials were performed in humans with one or more CVDRFs to evaluate the effects of different flavonoids. However, there are differences among different flavonoid and therefore, their effect on health cannot be generalized, and each type of flavonoid needs to be

61 studied. Thus, the effects of some of them are currently controversial, and some flavonoid classes have not been sufficiently studied to date. Thus, investigating how the consumption of specific flavonoids influences different conditions is useful to find new ways to prevent and treat diseases such as CVDs.

Figure 7 summarizes some of the known potential health benefits of dietary flavonoids.

Figure 7. Potential health benefits of dietary flavonoids. Source: (Del Rio D et al. 2013).(9)

3. 4 Mechanisms of action.

The first mechanism of action described for flavonoids was focused on their direct antioxidant effects.(56) However, the concentrations of these compounds in most tissues do not reach the minimum needed to have a significant effect in terms of scavenging free radicals.(56) Additionally, other molecular mechanisms of action have been identified such as the implication on cellular signalling pathways, regulating nuclear transcription factors and lipid metabolism, and modulating inflammatory mediators synthesis such as interleukins and TNFα.(72) In this sense, in vitro and in vivo as well as epidemiological and experimental studies highlighted the anti-inflammatory activity of flavonoids.(73)

While the molecular mechanisms continue to be clarified, the identified signalling pathways include AMP-activated protein kinase (AMPK), peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), peroxisome proliferator-activated receptor gamma (PPAR- γ) and nuclear factor-κB (NF-κB). Nevertheless, the mechanism of action by which several flavonoids exert beneficial effects remains unclear. On the other hand, there are many identified flavonoids to date, and the observed effects on molecular pathways for some of them are unlikely to be generalizable to others because of their differential structure and their different ways of acting on organism.(74)

CHAPTER 4. The flavonoid hesperidin.

In this section, the definition, characteristics, metabolism, bioavailability, and health effects of the flavonoid hesperidin will be presented. Moreover, we introduce the actual evidence of the effects of hesperidin on cardiovascular health since it is the focus of the present thesis.

The flavonoid hesperidin (hesperetin 7-O-rutinoside) is founded in glycosylated form in food, and when it is absorbed in the organism is transformed into aglycated form called hesperetin. In bloodstream, hesperetin can be conjugated into sulfate and glucuronides forms.(14) The chemical structures of hesperidin and hesperetin are represented in Figure 8.

Hesperidin is found in the peel of citrus fruits representing 90% of citrus flavonoids but a considerable amounts are found in their juices.(75) Moreover, hesperidin is the main citrus flavonoid of orange fruit and its juice, and orange and OJ are the most common citrus fruit products consumed in Europe.(17) Furthermore, the consumption of hesperidin through supplements or capsules is considered safe and harmless.(76)

Figure 8. Chemical structures of hesperidin form (A) founded in food, and hesperetin form (B) founded when hesperidin is absorbed. Source: (Xiong H et al. 2019).(77)

4. 1 Health effects of the flavonoid hesperidin.

In recent years, the flavonoid hesperidin has gained much attention in the phenolic compounds research due to its reported beneficial effects on health. Observational studies have reported that citrus fruit consumption is associated with a lower risk of acute coronary events.(78) Moreover, preclinical and clinical studies have demonstrated the possible therapeutic action of the flavonoid hesperidin on several diseases, such as psychiatric disorders, neurological disorders, carcinoma and CVDs, because of its anti-inflammatory, antioxidant, and other interesting properties.(79)

There is evidence of the beneficial effect of hesperidin on the cardiovascular system.(80) Figure 9 shows the possible beneficial effects of hesperidin consumption on several CVDRFs.

Figure 9. Effects of hesperidin consumption on cardiovascular disease risk factors in humans. Abbreviations: TG, triglycerides; SBP, systolic blood pressure; DBP, diastolic blood pressure (own source).

Lipid profile and adiposity: First, in animal models, hesperidin administration improves the lipid profile in rats with type 2 diabetes and hypercholesterolaemia. In this sense, it is known that dyslipidaemia is an important and significant risk factor for the development of atherosclerosis. Additionally, hesperidin administration increases high-density lipoprotein-cholesterol (HDL-c) levels and decreases TG, TC and LDL-c levels in metabolic syndrome rats.(81) Moreover, in type 2 diabetic rats, hesperidin decreased the plasma free fatty acids and plasma and hepatic TG levels after 5 weeks of daily consumption.(82) The improvement is produced because of the downregulation of the synthesis of very low density lipoprotein (VLDL) in hepatocytes, the inhibition of lipogenesis and the promotion of beta oxidation of fatty acids.(83) Moreover, hesperidin can suppress hepatic fatty acid synthase and glucose-6-phosphate dehydrogenase and decrease fatty acid oxidation and carnitine palmitoyl transferase activity.(84)

Second, similar results were obtained in human studies. In hypercholesterolaemic subjects, hesperidin consumption through OJ improved lipid profiles increasing HDL-c concentrations and decreasing the LDL-c/HDL-c ratio.(85,86) Thus, hesperidin might benefit atherosclerosis by reducing lipid levels because of its anti-lipid peroxidation and antioxidant properties.(87) However, unlike the studies in animals, in humans, there is controversy; some studies see similar results as those in animal studies, while other studies did not observe its effects.(88,89)

On the other hand, the alteration of adipose tissue and its dysfunction promotes the development of obesity, an independent risk factor for

CVDs, and hesperidin seems to be a possible therapeutic agent for obesity. It is known that the chemical structure of the flavonoid hesperidin is one of the most effective inhibitors of adipocyte formation since hesperidin can induce a decrease in TG concentration in preadipocytes.(90)

In animal models of obesity or metabolic syndrome, hesperidin reduces body weight and adipose tissue weight.(91,92) In contrast, in human studies the actual evidence is controversial. In some studies, daily hesperidin consumption reduces body weight in subjects with obesity or overweight and in hypercholesterolaemic subjects(93,94) and can increase adiponectin levels in patients with myocardial infarction.(86) However, other studies in humans did not show positive results for weight loss or obesity-related biomarkers.(95,96)

Blood pressure and endothelial function: As we previously detailed, high BP levels are one of the most important CVDRFs, and endothelial dysfunction is also an important contributor to the pathobiology of atherosclerotic CVD.(97) It has been demonstrated that hesperidin has an anti-hypertensive effect in renovascular hypertensive rats that involves the suppression of the renin-angiotensin system(98). Additionally, the anti-hypertensive effect of hesperidin was suggested to be mediated by the vascular NO synthase pathway and the reduction of oxidative stress by overexpression of NADPH oxidase, improving endothelial function in rats.(98) In vitro studies have demonstrated an increased production of NO in endothelial cells after hesperidin administration. Additionally, in vivo

68 studies in rats showed that hesperidin can prevent NO deficiency and hypertension, and the possible mechanism involved its anti- inflammatory and antioxidant effects.(99) Animal studies also showed the capacity of hesperidin to exert antioxidant activity. In this sense, NO is an important vasodilator produced by the vascular endothelium with the objective of regulating the vascular tone.(99) Thus, decreased NO production results in higher vascular resistance and increased BP levels. Therefore, hesperidin seems to improve this aspect.

In animal models and in vitro studies, the effects of hesperidin consumption on BP and endothelial function are clear; however, in human studies, the effects are controversial. In subjects with at least one CVDRF, the sustained hesperidin consumption decreases systolic and diastolic BP levels and improves the endothelial function.(71,93,100,101) Nevertheless, there are studies that did not show beneficial effects on BP and endothelial function.(102,103)

Glucose metabolism: The main complication of diabetes is CVDs, and there are several studies both in vitro and in vivo that have shown positive effects of hesperidin consumption on glucose homeostasis. Recently, in an in vitro study, both hesperidin and hesperetin inhibited the non-enzymatic glycation of proteins, the main reaction involved in the formation of advanced glycation end-products which has an important role in the pathogenesis of diabetes.(104) On the other hand, hesperidin can affect the gene expression of glucose-regulating enzymes, such as phosphoenolpyruvate carboxykinase and glucose-6- phosphatase, influencing glucose metabolism and glucose

69 regulation.(105) Additionally, hesperidin can decrease the blood glucose concentration by upregulating hepatic glucokinase and PPAR- γ and adipocyte GLUT4.(106) In human clinical trials, similar to the evidence for BP and endothelial function, there is no clear effect of hesperidin on glucose metabolism. To the best of our knowledge, to date no study has found a positive effect of hesperidin consumption on plasma glucose levels or insulin in humans.(101,103,107)

4. 3 Metabolism and bioavailability of the flavonoid hesperidin.

The absorption of flavonoid glycosides normally occurs in epithelial cells of the small intestine by the enzymes lactase phlorizin hydrolase and B-glycosidase, and the metabolites obtained are transported into the bloodstream.(9) However, bioavailability studies have shown that the flavonoid hesperidin is resistant to enzymes of the stomach and small intestine; therefore, hesperidin can arrive intact at the colon.(9) It is known that only 30% of hesperetin metabolites are absorbed in the small intestine and the other 70% are absorbed in the colon.(108) At the colon, hesperidin is converted to glucuronides form by alpha- ramnosidase and microbiota.(109) Then, the hesperidin molecule realizes the aglycone form called hesperetin. Hesperetin is released into the bloodstream in glucuronide form and sulfate conjugates.(110) Three of the most relevant metabolites of hesperidin founded in the organism are: hesperetin 7-O-B-D glucuronide, hesperetin 3-O-B-D glucuronide and hesperetin 7-O- sulfate.(111)

A schematic representation of hesperidin metabolization in the colon is represented in Figure 10.

The bioavailability of the flavanone hesperidin depends on the type of food matrix form in which it is ingested (juice, extract, capsules, etc.), the processing techniques, and the characteristics of the host, such as health status, age, sex, genetics and/or gut microbiota composition.(112) In this sense, it is known that the bioavailability of hesperidin is not different between orange fruit and OJ with higher doses of hesperidin. The similar bioavailability is likely due to the metabolism of flavanones being saturated when the intake exceeds a certain limit.(113) On the other hand, when hesperidin is consumed by capsules or supplements, the bioavailability is similar to hesperidin consumed by food or beverages.

Finally, it is known that subjects can be stratified into three categories depending on their capacity to absorb and excrete hesperidin metabolites: high, medium and low.(110) Therefore, as we previously mentioned, the bioavailability of the flavanone hesperidin depends on different factors that can be considered when the effects are studied.

Figure 10. Graphical representation of hesperidin metabolization in the colon. Source: (Mas-Capdevila A, et al. 2020).(112)

CHAPTER 5. Omics science: the future of research.

The word omics refers to an area of study in biological sciences that includes various disciplines including genomics, transcriptomics, proteomics, and metabolomics (Figure 11), as well as other emergent omics such as metagenomics, lipidomics, glycomics or foodomics. The ending -ome is used to describe the object of study of each field, such as the genome, transcriptome, proteome, or metabolome.(5)

The omics sciences involve the analysis of a large volume of data using bioinformatics tools to bring great advances in the basic knowledge of biological issues and the study of organisms and their functions, all through genetic tracing.(114) The omics approach provides a good option for hypothesis-generating experiments, as holistic approaches analyse all available data to describe a hypothesis in situations when no hypothesis is known or prescribed due to a lack of data.(5) In addition, as the omics sciences allow for study at the molecular level of the different elements that include biological systems (cells, tissues, etc.) in all their complexity, omics also allows for the study of the interactions and relationships that occur between the internal components of the individual and the external elements.

In this sense, omics applications can be used to prove the connections and interrelationships among the many scenarios in a complex physiologic state and to discover missing pieces in the current knowledge.(5)

Figure 11. Graphical representation of the main omics sciences (own source).

In addition, there are two types of analyses: nontargeted and targeted. Nontargeted approaches have the objective of detecting, identifying, and quantifying as many genes, proteins or metabolites in a biological sample as possible. In contrast, a targeted approach identifies a group of genes, proteins or metabolites previously selected.(115) Thus, the wide coverage of nontargeted approaches has the potential to identify new metabolic pathways and new disease biomarkers.

In this section of the present doctoral thesis, we focus on transcriptomics, metabolomics, and proteomic approaches.

5. 1 Transcriptomics approach.

The starting point of molecular biology represents the flow of genetic information from genes to the respective functions in cells and organisms. This process begins with the transcription of deoxyribonucleic acid (DNA), the genetic information repository. DNA is transcribed by the ribonucleic acid (RNA) polymerase enzymes into RNA and after that, a subset of RNA is translated into protein.

However, not all RNAs are translated into proteins. Some of them will have a structural function (such as rRNAs), develop a regulatory function (such as siRNAs) or be long non-coding RNAs (lncRNAs), that they do not be translated into proteins but demonstrate their capacity to play important roles in human diseases such as cardiovascular and cancer disorders. Therefore, the transcriptome is a set of RNA molecules present in a cell, a set of cells or in an organism. However, transcriptomics also provides important data regarding the content of cell non-coding RNAs, such as lncRNAs.(116)

Transcriptomics analysis is the study of the transcriptome, the complete set of RNA transcripts, that is produced by the genome under specific conditions in a specific cell, using high-throughput methods, such as microarray analysis. A microarray is a tool used to detect the expression of thousands of genes at the same time. DNA microarrays are microscope slides that are printed with thousands of tiny spots in defined positions, with each spot containing a known sequence or gene. Then, the unknown DNA molecules are cut into fragments by restriction endonucleases and fluorescent markers are attached to these DNA fragments, allowing the reaction with probes of the DNA microarray.(5)

Transcriptomic analysis is used in research and biomedicine for disease diagnosis, biomarker discovery, risk assessment of new drugs, etc. Moreover, transcriptomics analysis also allows the identification of pathways and metabolic changes produced because of environmental stresses such as diet. In this sense, transcriptome analysis is commonly used to compare samples that received different external conditions to

75 identify genes that are differentially expressed in distinct cell populations, or in response to different treatments. Additionally, it is commonly used to compare healthy and disease states to determine, for example, the mechanisms of pathogenesis.(116)

5. 2 Proteomics approach.

Proteomics, first used in 1996 by Marc Wilkins, is the study and characterization of the proteome. The proteome is defined as all the protein content of a cell, tissue or biological fluid in specific conditions, characterized by their localization, interactions, expression, structure and functions.(117) The proteome changes from time to time, from cell to cell and in response to external factors.

Proteomics analysis is one of the most important methods to determine gene function, although it is more complex than genomics. Proteins are effectors of biological function, and their levels not only depend on mRNA levels but also depend on the control and regulation of translation.(117) Therefore, proteomics is considered the most important set of data to characterize a biological system. In this sense, proteomics offers complementary information to genomics and transcriptomics needed to understand complex biochemical processes.(118) Furthermore, the understanding of protein expression, and thereby what and how biological processes are regulated at the protein level allows the understanding of the molecular basis for diseases and the option to shed light on disease prevention, diagnosis, and treatment.

There is diverse proteomics approaches and one of the most relevant tools to identify, characterize and quantify proteins on a large scale, is mass spectrometry (M-S), which allows the analysis of complex protein mixtures with high sensitivity.(119) M-S is a technique that ionizes all chemical compounds in a sample, resulting in charged molecules that are analysed depending on their mass-to-charge ratios. For the simple pre-separation of complex protein mixtures before M-S analysis, one- or two-dimensional polyacrylamide gel electrophoresis was often used in the past.(5) Currently, off gel-nanoLC-(Orbitap) MS/MS is used because it is more efficient in identifying novel proteins in various biological systems.

The identification of differentially expressed proteins in healthy and diseased subjects allow the discovery of new biomarkers in biomedical research. For example, recent work demonstrated the implication of several proteins in heart failure in subjects through proteomic analysis, allowing new plasma biomarker discovery for heart failure.(120) Moreover, as the application that will be realized in the present doctoral thesis, proteomics can provide insight into the effects of specific compounds, such as the flavonoid hesperidin, on the proteome to explore the molecular mechanisms involved in the cardioprotective effects of hesperidin.(121)

5. 3 Metabolomics approach.

Metabolomics emerged in the third millennium with the intention of measuring all the small molecule metabolites in a biological system, or at least most of them. For that reason, the improvement in instrument

77 technologies has improved the metabolomics analysis, most notably in the sensitivity and mass range of mass spectrometers.(122)

The metabolome is the global profile of metabolites in a biological sample. Metabolites are the natural intermediate products of metabolic reactions catalysed by enzymes that occur in cells, including an immense variety of endogenous small molecules such as amino acids, lipids, nucleic acids, sugars, fatty acids, etc., as well as exogenous chemicals such as pharmacological agents, toxins or xenobiotics. Thus, metabolomics is the study of metabolites in response to environmental stimuli and genetic alterations, giving a full-scale analysis of cellular and tissue metabolism and providing a more comprehensive understanding of biology.(122) Typically, metabolomics uses high-resolution analysis together with statistical analysis such as principal component analysis, to obtain an integrated picture of the metabolome. The most commonly used methods by which the metabolome is assessed are nuclear magnetic resonance spectroscopy and M-S.

The metabolic profiles obtained permit the development of diagnostic and/or prognostic tools that have the potential to significantly alter the management of CVDs.(123) For example, several works identified both urinary and plasma metabolites correlated with BP levels and hypertension, such as alanine or hippuric acid.(124) On the other hand, branched-chain amino acids (BCAA) levels have been linked with insulin resistance and type 2 diabetes mellitus, which increase CVD risk.(125)

For that reason, knowing the metabolome profile of individuals can offer translation from research to clinical practice to substantially affect the diagnosis, prognosis, and treatment of cardiovascular medicine. In nutritional interventions, a metabolomics approach can offer the possibility of identifying new biomarkers to determine the intake of specific compounds such as hesperidin and to relate it to the biological effects observed. Biomarkers are constituents present in biofluids such as blood and urine that can be used to indicate dietary exposure to compare it to the recommended or estimated intake. Moreover, in human intervention studies, biomarkers can indicate dietary changes and can be exposure biomarkers. Furthermore, metabolomics offers away to evaluate the contribution of dietary factors by associating circulating metabolites with global CVDs but also with specific CVDRFs such as BP.(126)

5. 4 Omics approach for the prevention and treatment of diseases.

The molecular measurements from large-scale omics data could be integrated into models of disease risk prediction adding valuable information to traditional clinical tests to better stratify patients into high- or low-risk groups for the potential of developing a disease such as CVD. However, currently, the cost of performing an omics analysis is high.

Nevertheless, the integration of different omics techniques is expected to become increasingly powerful for disease treatment and prognosis and has even been suggested to be useful in disease regression.(127)

The simultaneous integration of science omics was called systems biology in the past, and foodomics when it was applied to the food,(128) but nowadays it is called multi-omics (Figure 12). Multi-omics approach could be powerful in better understanding the mechanisms and pathways implicated to identify key drivers that have the largest contribution to diseases such as CVDs.

Figure 12. Workflow with integrated omics showing input datasets and results. Source: (Misra BB et al. 2018).(129)

3. HYPOTHESIS AND OBJECTIVES

HYPOTHESIS

Our hypothesis is that the intake of hesperidin, the main flavonoid of citrus fruits presents mainly in orange and OJ, might exert beneficial effects on CVDRFs, particularly reducing BP, in humans. Moreover, the mechanisms of action of hesperidin by which it could improve CVD can be elucidated through different omics approaches, such as transcriptomics, metabolomics, and proteomics, both in humans and in experimental animal models.

OBJECTIVES

The main objective is to evaluate the effects of the consumption of hesperidin in OJ on different CVDRFs on humans, particularly on BP, and to elucidate the possible mechanisms of action and the biological effects involved though the application of different omics approaches such as transcriptomics, metabolomics and proteomics, both in humans and animal models.

The following specific objectives have been set:

Objective 1: To summarize and evaluate the current scientific evidence from animal studies and human RCTs to determine the effects of hesperidin on cardiovascular risk biomarkers.

Objective 2: To assess the sustained and acute effects, and the sustained consumption influence on acute effects, of real-life doses of OJ and a

83 hesperidin-enriched dose on BP, PP, and cardiovascular risk biomarkers in pre- and stage-1 hypertensive individuals.

Objective 3: To determine whether the sustained and single dose consumption of hesperidin in OJ and EOJ can change the transcriptomic profile of PBMCs of subjects with pre- and stage 1 hypertension to elucidate possible mechanisms of action of the hesperidin and their role in CVD.

Objective 4: To determine new biomarkers of the biological effects of hesperidin in OJ applying nontargeted metabolomics approach in plasma/serum and urine samples after both single dose (6 hours) and sustained (12 weeks) consumption, validated through targeted metabolomics analyses of compliance biomarkers in subjects with pre- and stage 1 hypertension.

Objective 5: To determine the changes in proteomic profiles of kidney and heart tissues, as key organs involved in BP regulation and cardiovascular system, in healthy and metabolic syndrome rats after hesperidin supplementation to shed light on the hesperidin mechanism of action.

4. METHODS AND RESULTS

Project 1: Effects of hesperidin consumption on cardiovascular risk biomarkers: a systematic review of animal studies and human randomized clinical trials.

Lead Article Effects of hesperidin consumption on cardiovascular risk biomarkers: a systematic review of animal studies and human randomized clinical trials Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 L. Pla-Paga, J. Companys, L. Calderon-Perez, E. Llaurado, R. Sola, R. M. Valls, and A. Pedret

Context: The cardioprotective effects of the flavonoid hesperidin, which is present in citrus products, are controversial and unclear. This systematic review was conducted in accordance with the PRISMA 2015 guidelines. Objective: To evaluate the current evidence from animal and human clinical studies and thus determine whether the consumption of hesperidin exerts beneficial effects on cardiovascular risk factors. Data sources: PICOS (Population, Intervention, Comparison, Outcome, and Study Design) criteria defined the research question. Searches of the PubMed and Cochrane Plus databases were conducted and studies that met the inclusion criteria and were published in English in the last 15 years were included. Data extraction: The first author, year of publication, study design, characteristics of animals and humans, intervention groups, dose of hesperidin, route of administration, duration of the intervention, cardiovascular risk biomarkers assessed, and results observed were extracted from the included articles. Results: A total of 12 animal studies and 11 randomized clinical trials met the inclusion criteria. In the animal studies, the glucose, total and LDL cholesterol, and triglyceride levels decreased with chronic flavonoid consumption. In the human studies, endothelial function improved with flavonoid consumption, whereas no conclusive results were observed for the other biomarkers. Conclusions: Animal studies have revealed that hesperidin and hesperetin consumption reduces glucose levels and various lipid profile parameters. However, a definitive conclusion cannot be drawn from the existing human clinical trials. Further research is needed to confirm whether the findings observed in animal models can also be observed in humans. Systematic Review Registration: Prospero registration number CRD42018088942.

INTRODUCTION showed that ischemic heart disease and stroke caused 15 million deaths in 2015 worldwide.1 Cardiovascular diseases (CVDs) constitute the main Currently, there is a growing interest in identifying cause of mortality throughout the world.1 The latest sta- new bioactive compounds with healthy effects on tistical data from the World Health Organization CVDs, which can then be used to develop functional

Affiliation: L. Pla-Paga, J. Companys, L. Calderon-Perez, E. Llaurado, R. Sola, R. M. Valls, and A. Pedret are with the Eurecat, Centre Tecnologic de Catalunya, Unitat de Nutricio i Salut, Reus, Spain. L. Pla-Paga, J. Companys, L. Calderon-Perez, E. Llaurado, R. Sola, R. M. Valls, and A. Pedret are with the Universitat Rovira i Virgili, Facultat de Medicina i Cie`ncies de la Salut, Functional Nutrition, Oxidation and Cardiovascular Disease Group (NFOC-SALUT), Reus, Spain. L. Calderon-Perez and R. Sola are with the Hospital Universitari Sant Joan, Reus, Spain. Correspondence: R.M. Valls, Facultat de Medicina i Cie`ncies de la Salut, Universitat Rovira i Virgili. Carrer Sant Llorenc¸ 21, 43201, Reus, Spain. E-mail: [email protected]; R. Sola, Facultat de Medicina i Cie`ncies de la Salut, Universitat Rovira i Virgili. Carrer Sant Llorenc¸ 21, 43201, Reus, Spain. E-mail: [email protected] Key words: cardiovascular risk biomarkers, citrus flavonoids, glucose, hesperidin, hesperetin, lipid profile.

VC The Author(s) 2019. Published by Oxford University Press on behalf of the International Life Sciences Institute. All rights reserved. For permissions, please e-mail: [email protected]. doi: 10.1093/nutrit/nuz036 Nutrition ReviewsVR Vol. 77(12):845–864 845 foods, and phenolic compounds have gained much in- associated with a lower risk of acute coronary terest in this field of research. Polyphenols are second- events.17,18 However, the findings from human ran- ary metabolites of plants, and more than 8 000 different domized clinical trials (RCTs) are not consistent: some types exist, which can be classified into different groups studies have found that daily consumption of orange depending on the number of phenolic rings they con- juice decreases systolic blood pressure (SBP) and dia- tain and the type of substituent attached to the rings.2 stolic blood pressure (DBP),19 and increases the total Polyphenols are divided into two large families: flavo- plasma antioxidant capacity or decreases lipid peroxida- 20 noids and non-flavonoids. Flavonoids are the most tion, but others have not reported any beneficial Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 abundant type in plants, and the main subclasses in- effects on blood pressure or the lipid profile after hes- clude flavonols, flavones, isoflavones, flavanones, antho- peridin consumption.21,22 To the best of our knowledge, cyanidins, flavan-3-ols, and dihydrochalcones.2 the current scientific evidence on the effects of hesperi- Flavonoids can be found in many commonly consumed din on cardiovascular risk biomarkers obtained from fruits and vegetables, and numerous studies have shown animal studies and human RCTs has not been systemat- their benefits for the prevention and treatment of differ- ically reviewed, and thus, no conclusive remarks can be ent pathologies.3–5 In recent years, citrus flavonoids, drawn. which are present in different citrus fruits, particularly Therefore, the present systematic review aimed to in orange juice, have gained the attention of the food determine whether hesperidin consumption might exert industry because they may exert beneficial effects on beneficial effects on cardiovascular risk biomarkers. different cardiovascular risk factors (CVRFs)6 and be- The objective was to summarize and evaluate the cur- cause orange juice is one of the most consumed bever- rent scientific evidence from animal studies and human ages throughout the world.7 In European adults, the RCTs to determine the effects of hesperidin on cardio- mean flavonoid intake is 428 mg/day.8 vascular risk biomarkers. The main citrus flavonoid of orange fruit and orange juice is hesperidin, which is found in greater quan- METHODS tities in the peel and represents 90% of citrus flavonoids.9 Hesperidin (hesperetin-7-O-rutinoside) is a This systematic review was conducted according to flavanone glycoside and the dietary form of the agly- the PRISMA 201523 (Preferred Reporting Items for cone hesperitin.6 Normally, the absorption of flavonoid Systematic reviews and Meta-Analysis) guidelines glycosides such as hesperidin occurs in epithelial cells and was registered with PROSPERO on February 20, in the small intestine and is facilitated by the enzymes 2018, under the ID number CRD42018088942. The lactase phlorizin hydrolase or cytosolic b-glucosidase, protocol can be accessed at http://www.crd.york.ac. resulting in the separation of the aglycone and its trans- uk/PROSPERO/display_record.php? ID¼CRD42018 portation into the bloodstream.2 Then, the metabolites 088942. are transported to the liver for phase II metabolism, and they can be recycled by the enterohepatic recirculation Eligibility criteria in the small intestine. However, bioavailability studies show that only 30% of hesperetin metabolites are Animal studies and RCTs were eligible for the system- absorbed in the small intestine and the other 70% are atic review in accordance with the review’s PICOS crite- absorbed in the colon,2 via microbiota and alpha- ria. The complete PICOS criteria for inclusion and rhamnosidase activity,10 where the hesperidin is con- exclusion of studies are described in Table 1. verted to glucuronides. In-vitro studies have revealed that hesperidin stimulates the production of nitric oxide Information sources, search strategy, and study (NO) in endothelial cells,11,12 inhibits the secretion of selection endothelin-112 and inhibits platelet activity by inhibiting the activities of specific phospholipases and cycloox- A literature search of the PubMed (https://www.ncbi. ygenase-1.13 Animal studies have shown that hesperidin nlm.nih.gov/pubmed/) and Cochrane Plus (www.biblio- exhibits antioxidant capacity and endothelial protection tecaco-gov/pubmed) databases was performed using against reactive oxygen species in spontaneously hyper- medical subject headings (MeSH). The complete search tensive rats, and improves hyperlipidemia and hyper- strategy is shown in Table 2. The literature search was glycemia in diabetic rats.14 Conversely, other animal restricted to English-language articles published be- studies have not found that hesperidin exerts beneficial tween January 2003 and January 2018. effects on glucose or insulin levels, lipid profile, or To ensure the accurate identification of eligible blood pressure.15,16 In contrast, several observational studies, a two-step selection process was used. To con- studies have shown that citrus fruit consumption is firm the eligibility of the included articles, the titles and

846 Nutrition ReviewsVR Vol. 77(12):845–864 Table 1 PICOS criteria for inclusion and exclusion of studies Parameter Inclusion criteria Exclusion criteria For the animal studies Participants Rats or mice with at least one CVRF (obesity, dyslipidemia, Studies performed on animal models that were not hypertension, diabetes, or metabolic syndrome) rats or mice Intervention Some type of intervention based on hesperidin Combination of different classes of phenolic compounds (other than citrus flavonoids) and combination with other nutrients, components, or drugs

(vitamin C, caffeine, or hypertension drugs) Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 Comparisons Different doses of hesperidin and/or hesperidin consumption and non-consumption Outcomes Studies that assessed the effects of hesperidin on biomarkers or risk factors related to CVDs: anthropometric parameters, vascular parameters, glucose and insulin levels, lipid profile and coagulation, inflammation and oxidation biomarkers Study design Randomized and non-randomized, acute and chronic fol- Studies published before January 2003 and in any low-up, published in English language other than English For the RCTs Participants Humans of all races, ages, and genders with at least one Humans with no CVRF CVRF (obesity, dyslipidemia, hypertension, diabetes, or metabolic syndrome) Intervention Some type of nutritional intervention based on the con- Combination of different classes of phenolic com- sumption of hesperidin from food, drink, or supplement pounds (other than citrus flavonoids) and combination with other nutrients, components, or drugs (vitamin C, caffeine, or hypertension drugs) Comparisons Different doses of hesperidin and/or hesperidin consumption and non-consumption Outcomes Studies that assessed the effects of hesperidin consumption on biomarkers or risk factors related to CVD: anthropometric parameters, vascular parameters, glucose and insulin levels, lipid profile and coagulation, inflammation and oxidation biomarkers Study design Randomized controlled clinical trials, parallel and crossover Reviews, expert opinion, comments, letter to editor, design, acute and chronic follow-up, published in case reports, conference reports, observational English studies, animal studies, and studies published before January 2003 and in any language other than English Abbreviations: CVDs, cardiovascular diseases; CVRF, cardiovascular risk factor; RCTs, randomized clinical trials. abstracts of the studies identified using the search strat- groups, dose of hesperidin, route of administration, du- egy were screened independently by two authors (LP-P ration of the intervention, cardiovascular risk (CVR) and JC). The full text of the potentially eligible studies biomarkers assessed, and results observed. was then retrieved and independently assessed for eligi- The following data were extracted from the RCTs: bility by the same two authors. Any disagreement be- first author, year of publication, study population, pop- tween the authors over the eligibility of a study was ulation age and health status, characteristics of the nu- resolved through discussion with a third author (LC-P). tritional intervention, dose of hesperidin, consumption matrix, duration of the intervention, method used to Data collection and extraction confirm compliance with the intervention, CVR biomarkers assessed, and results observed. From the total number of articles identified by assign- ing appropriate MeSH terms, any duplicate articles Study quality and risk of bias in the individual studies within and between the databases were removed. The remaining articles were assessed primarily according to Assessments of the quality and possible risks of bias in their title and abstract, and then according to their full each RCT included in the present systematic review text, and those studies that did not meet the eligibility were performed using Review Manager software criteria were removed. (RevMan; version 5.3), a tool provided by the Cochrane The following data were extracted from the in- Collaboration. The following items were included in the cluded animal studies: first author, year of publication, assessments: random sequence generation, allocation study design, characteristics of the animals, intervention concealment, blinding of participants and personnel,

Nutrition ReviewsVR Vol. 77(12):845–864 847 Table 2 Search strategy and MeSH terms used For the animal studies For the RCTs Search strategy: Search strategy: -Electronic databases: PubMed and Cochrane Plus -Electronic databases: PubMed and Cochrane Plus -Publication dates: January 2003 – January 2018 -Publication dates: January 2003 – January 2018 -Species: Other animals -Species: Humans

MeSH terms: MeSH terms: hesperidin orange juice Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 hesperetin orange polyphenols and citrus flavonoids blood pressure citrus flavanones endothelial function hesperidin blood cholesterol hesperetin high density lipoprotein and low density lipoprotein blood pressure apolipoprotein A1 hypertension apolipoprotein B100 endothelial function triglycerides blood cholesterol plasma no esterified reactive protein high density lipoprotein glucose low density lipoprotein insulin resistance apolipoprotein A1 diabetes apolipoprotein B100 C-reactive protein triglycerides IL-6 plasma no esterified reactive protein IL-18 glucose nitrates and nitrites insulin resistance platelet aggregation diabetes endothelin IL-6 soluble intercellular adhesion molecule-1 IL-18 soluble vascular cell adhesion molecule-1 nitrates and nitrites platelet aggregation E-selection endothelin serum amyloid A soluble intercellular adhesion molecule-1 oxidized low density lipoprotein soluble vascular cell adhesion molecule-1 urinary creatinine E-selection oxidative stress serum amyloid A nitric oxide oxidized low density lipoprotein homocysteine urinary creatinine nitrotyrosine oxidative stress plasminogen activator inhibitor-1 nitric oxide von Willebrand factor homocysteine fibrinogen nitrotyrosine body mass index plasminogen activator inhibitor-1 body weight von Willebrand factor obesity fibrinogen overweight body mass index body weight obesity overweight atherosclerosis cardiovascular risk factors

Abbreviation: RCTs, randomized clinical trials.

blinding of outcome assessments, incomplete outcome RESULTS data, selective reporting, and other biases. The risk of bias in each study was classified as “low,” “unclear,” or Animal studies “high.” Two authors (LP-P and JC) evaluated the risk of bias in the RCTs, and any disagreement between them Study selection. A total of 698 articles were identified over the risk of bias of a study was resolved through dis- from the two databases (643 in PubMed and 55 in cussion with a third author (LC-P). Cochrane Plus). Of these, 367 duplicate articles were

848 Nutrition ReviewsVR Vol. 77(12):845–864 Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020

Figure 1 Flow diagram of the literature search process for animal studies. removed and 292 of the remaining 331 articles were ex- that was not administered flavanone. In 9 of the studies, cluded because they did not meet the inclusion criteria, hesperidin was orally administered,15,21,22,24–29 while in 7 were excluded because they were review articles, and 2 of the remaining 3 studies hesperidin was adminis- 6 were excluded because no full text was available. As a tered by gavage,30,31 and in the other study hesperidin result, 12 articles were included in the systematic re- was administered intravenously.32 The doses of hesperi- view. Figure 1 shows the study selection process for the din ranged from 5 mg/kg of body weight/day to 200 animal studies included in the review. mg/kg of body weight/day in 10 studies and from 0.08% to 4.60% of the total calorie intake in the other 2 studies. Study characteristics. Table 3 shows the general charac- The duration of the intervention ranged from 7 days to teristics of the 12 animal studies included in the system- 24 weeks. All the animals had at least one CVRF, such atic review. Further details of each study are presented as hypertension, myocardial ischemia, systemic inflam- in Table S1 in the Supporting Information online. The mation, hypercholesterolemia, and type 2 diabetes. The 12 studies included in the systematic review were con- sample size ranged from 4 to 16 animals in each group, trolled animal studies involving an intervention group and of the 12 studies, 8 were performed on rats and 4 that was administered flavanone and a control group on mice.

Nutrition ReviewsVR Vol. 77(12):845–864 849 850 Table 3 Characteristics and results of the animal studies included in the systematic review (n ¼ 12) Author, year, Experimental Groups (n) Dose of Route Duration Results reference animal flavanone Anthropometric Vascular Glucose and Lipid profile INFL. OXID. parameters parameters insulin levels BIOM BIOM BW WG VF SBP DBP GL INS TC LDL-c HDL-c TG IL-6 NO Iskender et al Type 2 diabetic Control group 0mg Orally 15 d NS – – – – # ––––––– (2017)24 rats (n¼10) vs Hesperidin group 100 mg/kg (n¼10) BW/d Dobiasetal Spontaneously Control group 0mg Orally 4 wk NS – – NS – – – – – – – – – (2016)25 hypertensive (n¼13) rats vs Hesperidin group 50 mg/kg (n¼13) BW/d Ferreira et al Mice with Control group 0mg Orally 4 wk NS NS NS – – NS – NS # NS NS # – (2016)15 systemic (n¼10) inflammation vs Hesperidin group 100 mg/kg (n¼10) BW/d Jia et al Type 2 diabetic Control group 0mg By gavage 6 wk NS – – – – # – # –– # –– (2015)31 mice (n¼10) vs Neohesperidin 50 mg/kg group (n¼10) BW/d Yamamoto et Hypertensive rats Control group 0mg Intravenous 3 min –––# with NS – – – – – – – – al (2013)32 (n¼4) (acute s hesperetin vs tudy) Hesperetin group 5 mg/kg BW/d (n¼4) vs Hesperetin-7-O-b- 5 mg/kg BW/d # with D-glucuronide HESPT7G urto Reviews Nutrition group (n¼4) vs Hesperetin-30- 5 mg/kg BW/d O-b-D- glucuronide V R group (n¼4) o.77(12):845–864 Vol. Kumar et al Type 2 diabetic Control group 0mg By gavage 24 wk – – – – – # ––––––– (2012)30 rats (n¼16) vs Hesperetin group 200 mg/kg (n¼16) BW/d

(continued) Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 November 03 on Dent Carly access, site OUP by https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 from Downloaded urto Reviews Nutrition Table 3 Continued Author, year, Experimental Groups (n) Dose of Route Duration Results reference animal flavanone Anthropometric Vascular Glucose and Lipid profile INFL. OXID. parameters parameters insulin levels BIOM BIOM V R o.77(12):845–864 Vol. BW WG VF SBP DBP GL INS TC LDL-c HDL-c TG IL-6 NO Mahmoud et Type 2 diabetic Control group 0mg Orally 30 d – – – – – #"–– – –– # al (2012)26 rats (n¼6) vs Hesperidin group 50 mg/kg (n¼6) BW/d Selvaraj and Rats with myo- Control group 0mg Orally 7 d – – – – – – – ## " #–– Pugalendi cardial (n¼6) (2012)22 ischemia vs Hesperidin group 200 mg/kg (n¼6) BW/d Wang et al Hypercholestero- Control group 0% Orally 12 wk NS – – NS NS – – – – – – – – (2011)16 lemic rats (n¼15) vs Hesperidin (n¼15) 0.08% TCD/d Akiyama et al Type 2 diabetic Control group 0% TCD/d Orally 4 wk # ––– – # # # –– # –– (2009)27 rats (n¼6) with 4.6% with 4.6% with 4.6% vs Hesperidin group 1% TCD/d 1(n¼6) vs Hesperidin group 4.6% TCD/d 2(n¼6) Jung et al Type 2 diabetic Control group 0mg Orally 5 wk – – – – – # – # –NS# –– (2006)28 mice (n¼10) vs Hesperidin group 200 mg/kg (n¼10) BW/d Jung et al Type 2 diabetic Control group 0mg Orally 5 wk – – – – – – " –– – –– – (2004)29 mice (n¼10) vs Hesperidin group 200 mg/kg (n¼10) BW/d -, parameter not evaluated; #, significant decrease in intervention group vs control group; ", significant increase in intervention group vs control group. Abbreviations: BW, body weight; DBP, diastolic blood pressure; GL, glucose; HDL-c, high density lipoprotein; IL-6, interleukin-6; INFL. BIOM, inflammation biomarkers; INS, insulin; LDL-c, low-density lipoprotein; NO, nitric oxide; NS, no significant differences between intervention group and control group; OXID. BIOM, oxidation biomarkers; SBP, systolic blood pressure; TC, total cholesterol; TCD, total calorie diet; TG, triglycerides; VF, visceral fat; WG, weight gain.

851 Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 November 03 on Dent Carly access, site OUP by https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 from Downloaded Results for anthropometric parameters. The effect of hes- 2 diabetic rats by 1.61 mmol/L, compared with the con- peridin consumption on body weight was evaluated in 6 trol group. In addition, Jung et al28 reported that the studies.15,21,24,25,27,30 Of these, 4 studies reported no sig- oral consumption of 200 mg/kg of body weight/day of nificant changes,15,16,24,25 1 study reported a significant hesperidin for 5 weeks significantly lowered blood glu- decrease,27 and 1 study did not specify the outcome.30 cose levels in type 2 diabetic mice by 7.84 mmol/L, Akiyama et al27 reported that the administration of a compared with the control group. daily oral dose of hesperetin of 4.60% of total calorie in- The effect of hesperidin on serum insulin levels was 26,27,29 take to type 2 diabetic rats for 4 weeks prevented the evaluated in 3 studies, of which 2 reported signifi- Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 weight gain, of 13.56 g, observed in the control group. cant increases in insulin levels26,29 and 1 reported a In another study that examined the effect of hesperidin significant decrease.27 Mahmoud et al26 reported that consumption on visceral fat,15 no significant changes the oral consumption of 50 mg/kg of body weight/day were observed. of hesperidin for 30 days significantly raised insulin levels in type 2 diabetic rats by 6.05 mU/mL, compared with the control group. Jung et al29 found a significant Results for vascular parameters. The effect of hesperidin increase of 18.13 mU/mL in the insulin levels of type 2 on SBP was evaluated in 3 studies21,25,32: 2 of these stud- diabetic mice after 5 weeks of oral consumption of 200 ies reported no significant changes21,25 and 1 study mg/kg of body weight/day of hesperidin, compared reported a significant decrease in SBP.32 Yamamoto et with the control group. Moreover, Akiyama et al27 al32 reported that intravenous administration of an reported that oral daily consumption of hesperidin at a acute dose of 5 mg/kg of body weight of hesperetin to dose of 4.60% of total calorie intake for 5 weeks of inter- hypertensive rats significantly decreased SBP by 9.90 6 vention significantly lowered insulin levels in type 2 di- 1.70 mmHg, compared with the control group. The abetic rats by 90.64 mU/mL, compared with the control same study32 also reported that an acute dose of 5 mg/ group. kg of body weight of hesperetin-7-O-b-D-glucuronide significantly decreased SBP by 8.70 6 0.80 mmHg, Results for lipid profile. The effect of hesperidin con- compared with the control group. The effect of hesperi- sumption on total cholesterol (TC) levels was evaluated din on DBP was evaluated in 2 studies,21,32 but no sig- in 4 studies.15,22,27,28 Of these, 3 reported significant nificant changes were observed. decreases in TC levels22,27,28 and 1 study found no significant changes.15 Selvaraj and Pugalendi22 observed Results for glucose and insulin levels. The effect of hes- that the oral consumption of 200 mg/kg of body peridin on blood glucose was evaluated in 7 stud- weight/day of hesperidin for 7 days significantly low- ies.15,24,26–28,30,31 Six of these studies reported decreases ered TC levels in rats with myocardial ischemia by 0.40 in blood glucose24,26–28,30,31 and 1 study found no sig- mmol/L, compared with the control group. Akiyama et nificant changes.15 Iskender et al24 reported that the al27 noted that daily consumption of hesperetin at a oral consumption of 100 mg/kg of body weight/day of dose of 1% and 4.60% of total calorie intake for 4 weeks hesperidin for 15 days significantly lowered blood glu- significantly lowered TC levels in type 2 diabetic rats by cose levels in type 2 diabetic rats by 9.25 mmol/L, com- 1.71 mmol/L and 2.51 mmol/L, respectively, compared pared with the control group. Jia et al31 observed that with the control group. Moreover, Jung et al28 reported the consumption of 50 mg/kg of body weight/day of that the oral consumption of 200 mg/kg of body neohesperidin (derived from hesperidin) by gavage for weight/day of hesperidin for 5 weeks significantly low- 6 weeks significantly lowered blood glucose levels in ered TC levels in type 2 diabetic mice by 0.81 mmol/L, type 2 diabetic mice by 7.73 mmol/L, compared with compared with the control group. the control group. Kumar et al30 found that the con- The effect of hesperidin consumption on high-den- sumption of 200 mg/kg of body weight/day of hespere- sity lipoprotein cholesterol (HDL-c) levels was evalu- tin by gavage for 24 weeks significantly lowered blood ated in 3 studies.15,22,28 Two of these studies reported glucose levels in type 2 diabetic rats by 5.99 mmol/L, no significant changes in HDL-c15,28 and the other compared with the control group. Mahmoud et al26 study reported a significant increase.22 Selvaraj and detected significant reductions – of 9.49 mmol/L – in Pugalendi22 reported that the oral consumption of 200 the blood glucose levels of type 2 diabetic rats after oral mg/kg of body weight/day of hesperidin for 7 days sig- consumption of 50 mg/kg of body weight/day of hes- nificantly increased HDL-c levels in rats with myocar- peridin for 30 days, compared with the control group. dial ischemia by 0.34 mmol/L, compared with the Akiyama et al27 found that daily consumption of hes- control group. peretin at a dose of 4.60% of total calorie intake for 4 The effect of hesperidin consumption on low-den- weeks significantly lowered blood glucose levels in type sity lipoprotein cholesterol (LDL-c) levels was assessed

852 Nutrition ReviewsVR Vol. 77(12):845–864 in 2 studies,15,22 and significant decreases were observed because they did not meet the inclusion criteria. Thus, in both studies. Ferreira et al15 observed that the oral 11 articles were included in the systematic review. consumption of 100 mg/kg of body weight/day of hes- Figure 2 shows the study selection process for the RCTs peridin for 15 days significantly lowered LDL-c levels in included in the review. mice with systemic inflammation by 0.29 mmol/L, compared with the control group. In addition, Selvaraj and Study characteristics. Tables 4 to 6 show the characteris- Pugalendi22 observed that the oral consumption of 200 tics of the RCT studies included in this systematic re- mg/kg of body weight/day of hesperidin for 7 days sig- view. Further details of each study are presented in Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 nificantly decreased LDL-c by 0.67 mmol/L, compared Table S2 in the Supporting Information online. The 11 with the control group. studies included in this review were RCTs involving The effect of hesperidin consumption on triglycer- some type of nutritional intervention. In fact, the inter- ide (TG) levels was evaluated in 5 studies.15,22,27,28,31 ventions in 3 of the 11 included RCTs consisted of sup- Four of these studies reported significant plementation with a placebo capsule or a hesperidin decreases22,27,28,31 and the other study reported no sig- capsule,11,33,34 whereas those in the 3 other studies con- nificant changes.15 Jia et al31 noted that the consump- sisted of the administration of a control drink (CD) or tion of 50 mg/kg of body weight/day of neohesperidin orange juice (OJ).35–37 In addition, the interventions in by gavage for 6 weeks significantly lowered TG levels in 2 other studies involved the consumption of different type 2 diabetic mice by 2.05 mmol/L, compared with drinks with different hesperidin concentrations,38,39 the control group. In rats with myocardial ischemia, whereas those in 2 and 1 of the remaining RCTs con- Selvaraj and Pugalendi22 observed that the oral consisted of no product intervention vs OJ administra- sumption of 200 mg/kg of body weight/day of hesperi- tion40,41 and supplementation with a placebo or din for 7 days significantly lowered TG levels by 0.18 hesperidin capsule or consumption of OJ, mmol/L, compared with the control group. Akiyama et respectively.42 al27 noted that daily consumption of hesperetin at a Four of the studies comprised a parallel de- dose of 1% and 4.60% of total calorie intake for 4 weeks sign,33,34,40,41 and the other seven comprised crossover lowered TG levels in type 2 diabetic rats by 0.66 mol/L designs.11,35–39,42 Ten of the included RCTs involved a and 0.91 mmol/L, respectively, compared with the con- long-term follow-up, and one of these also involved a trol group. Lastly, Jung et al28 stated that the oral con- short-term follow-up. The other RCT involved only a sumption of 200 mg/kg of body weight/day of short-term follow-up. The duration of the intervention hesperidin for 5 weeks of intervention lowered TG levin the long-term studies ranged from 1.5 to 13 weeks, els in type 2 diabetic mice by 1.74 mmol/L, compared and the duration in the short-term studies ranged from with the control group. 4 to 5 hours. Nine of the studies were conducted with European populations, and the other 2 investigated Results for inflammation biomarkers. The effect of hes- South American populations. The sample sizes ranged peridin on interleukin-6 (IL-6) levels was evaluated in a from 22 to 194 subjects, and the ages of the subjects study by Ferreira et al.15 Using a mouse model of sys- ranged from 18 to 69 years. All the subjects had at least temic inflammation, this study reported a significant one CVRF, such as dyslipidemia, overweight, obesity, decrease of 58.64 pg/mL after the oral consumption of and/or metabolic syndrome. The methods used to con- 100 mg/kg of body weight/day of hesperidin for 4 firm intervention compliance involved keeping 3- or 5- weeks, compared with the control group. day food records, maintaining 24-hour dietary records, returning all used and unused capsule boxes, and self- Results for oxidation biomarkers. The effect of hesperi- reporting. din on nitric oxide levels was evaluated in a study by Mahmoud et al.26 The study reported a significant de- Assessment of the quality and risk of bias. The risk of crease of 5.08 mg/dL after the oral consumption of 50 bias in each individual RCT is detailed in Figure 3. Six mg/of body weight/day of hesperidin for 30 days in type of the 11 RCTs used an adequate random sequence gen- 2 diabetic rats, compared with the control group. erator; 3 studies incorporated adequate allocation concealment; 5 studies performed adequate blinding of the Human randomized controlled trials participants, personnel, and outcome assessment; 9 studies presented completed data; and 6 studies pre- Study selection. A total of 1917 articles were identified sented their study protocol with all the reported out- from the searches of the two databases (1 495 in comes. Regarding other types of bias, potential conflicts PubMed and 422 in Cochrane Plus). Of these, 1 486 du- of interest were considered, and 8 studies reported a plicate articles were removed and 393 were excluded lack of conflicts of interest.

Nutrition ReviewsVR Vol. 77(12):845–864 853 Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020

Figure 2 Flow diagram of the literature search process for randomized clinical trials.

Effects of chronic hesperidin consumption on with basal levels. No differences between the different cardiovascular risk biomarkers. hesperidin concentrations were observed. Aptekmann and Cesar40 noted that the consumption of 54.60 mg/ Results for anthropometric parameters. The characteris- day of hesperetin in 500 mL/day of OJ for 13 weeks of tics of the long-term RCTs included in this review in re- intervention significantly reduced the body weight of lation to anthropometric parameters are detailed in hypercholesterolemic subjects by 1 kg, compared with Table 4. basal levels. No significant differences were observed The effect of hesperidin consumption on body between the intervention and control groups. weight was evaluated in 3 studies.34,39,40 Of these, 2 39,40 The effect of hesperidin consumption on the body studies reported significant decreases, and 1 study 11,34,39–41 34 39 mass index (BMI) was evaluated in 5 studies :3 found no significant changes. Rangel-Huerta et al 11,34,41 of these studies reported no significant changes observed that the consumption of 237 mg/day or 582.50 39,40 and 2 studies found significant decreases. Rangel- mg/day of hesperidin in 500 mL/day of OJ for 12 weeks 39 Huerta et al observed that the consumption of 237 reduced the body weight of overweight or obese sub- mg/day and 582.50 mg/day of hesperidin in 500 mL/day jects by 1.30 kg and 1.80 kg, respectively, compared

854 Nutrition ReviewsVR Vol. 77(12):845–864 urto Reviews Nutrition Table 4 Characteristics of the long-term randomized clinical trials included in the systematic review with results for anthropometric and vascular parameters and glucose and insulin levels (n ¼ 8) Author, year, Sample Cardiovascular Nutritional intervention Results reference size risk factors Groups Flavanone dose Productdose Duration Anthropometric Vascular parameters Glucose and insulin levels parameters V R BW, kg BMI, kg/m2 BF, % SBP, mm Hg DBP, mm Hg FMD, % GTN, % GLUC, INS, mU/mL QUICKI o.77(12):845–864 Vol. mmol/L Salden et al n¼65 Overweight Placebo capsule 0 mg/d 500 mg/d 6 wk – – – NS/NS NS/NS NS/NS – NS/NS NS/NS NS/NS (2016)33 or obesity vs (NS) (NS) (NS) (NS) (NS) (NS) Hesperidin capsule 450 mg/d 500 mg/d Kean et al n¼37 Overweight Low hesperidin 64 mg/d 500 mL/d 8 wk – – – NS/NS NS/NS ––––– (2015)38 drink (NS) (NS) vs High hesperidin drink 549 mg/d 500 mL/d Constans et al n¼25 Hypercholesterolemia Control drink 0 mg/d 600 mL/d 4 wk – – – – – – – NS/NS(NS) – – (2015)35 vs Orange juice 213 mg/d 600 mL/d (hesperidin) Rangel-Huerta n¼100 Overweight or obesity Orange juice 237 mg/d 500 mL/d 12 wk #1.3/#1.8 #0.5/#0.7 – #4.00/NS #3.00/NS ––"0.3/"0.2 #1.2/NS – et al (2015)39 (hesperidin) (NS) (NS) (NS) (NS) (P <0.05)* (P <0.05)* vs Enriched orange juice 582.5 mg/d 500 mL/d (hesperidin) Buscemi et al n¼31 Metabolic syndrome Control drink 0 mg/d 500 mL/d 1.5 wk – – – – – NS/"2.2 NS/NS(NS) – – – (2012)37 vs (P <0.05)* Orange juice 159.5 mg/d 500 mL/d (hesperidin) Rizza et al n¼24 Metabolic syndrome Placebo capsule 0 mg/d 1 capsule/d 3 wk – NS/NS – NS/NS NS/NS NS/"2.5 NS/NS NS/NS NS/NS NS/NS (2011)11 vs (NS) (NS) (NS) (p¼0.02)* (NS) (NS) (NS) (NS) Hesperidin capsule 500 mg/d 1 capsule/d Morand et al n¼23 Overweight Placebo capsule with 0 mg/d 1 capsule and 4 wk – – – NS/NS/NS NS/#5.3/#4.5 – – NS/NS/NS NS/NS/NS – (2011)42 control drink 500 mL/d (NS) (P <0.023)* (NS) (NS) vs Hesperidin capsule 292 mg/d 1 capsule and with control drink 500 mL/d vs 500 mL/d Orange juice 292 mg/d (hesperidin) Demonty et al n¼194 Overweight Placebo capsule 0 mg/d 4 capsules/d 4 wk NS/NS NS/NS –– – – – – – – (2010)34 Hypercholesterolemia vs (NS) (NS) Hesperidin capsule 800 mg/d 4 capsules/d Aptekmann and n¼26 Overweight No intervention 0 mg/d 0 mL/d 13 wk #1.8/#1.0 #0.7/#0.3 #5.5/#4.3 –– – –––– Cesar (2010)40 vs (NS) (NS) (NS) Orange juice 54.6 mg/d 500 mL/d (hesperetin) Cesar et al n¼22 Hyperchole No intervention 0 mg/d 0 mL/d 8.5 wk – NS/NS –– – – – – – – (2010)41 sterolemia vs (NS) Orange juice 42 mg/d 750 mL/d (hesperetin) -, parameter not evaluated; #, significant decrease between basal and final value within each group; ", significant increase between basal and final value within each group; *, significant differences between groups. Abbreviations: BF, body fat; BMI, body mass index; BW, body weight; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; GLUC, glucose; GTN, glyceryl-nitrate dilation of the brachial artery; INS, insulin; NS, no significant differences between basal and final within each group; (NS); no significant differences between intervention group and control group; QUICKI, quantitative insulin-sensitivity check index; SBP, systolic blood pressure.

855 Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 November 03 on Dent Carly access, site OUP by https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 from Downloaded Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020

Figure 3 Risk of bias graph (A) and summary (B) of the randomized clinical trials included. 1 indicates a low risk of bias, - indicates a high risk of bias, and ? indicates an unclear risk. of OJ for 12 weeks reduced the BMI of overweight or observed.40 Specifically, Aptekmann and Cesar40 reported obese subjects by 0.50 kg/m2 and 0.70 kg/m2, respectively, that 54.60 mg/day of hesperetin in 500 mL/day of OJ sig- compared with basal levels. No differences between the nificantly reduced the body fat of hypercholesterolemic different hesperidin concentrations were observed. subjects by 4.30% after 13 weeks of intervention, com- Aptekmann and Cesar40 reported that the consumption of pared with basal levels. No significant differences were ob- 54.60 mg/day of hesperetin in 500 mL/day of OJ signifi- served between the intervention and control groups. cantly reduced the BMI of hypercholesterolemic subjects 2 by 0.30 kg/m after 13 weeks of intervention, compared Results for vascular parameters. The characteristics of with basal levels. No significant differences were observed the long-term RCTs included in this review in relation between the intervention and control groups. to vascular parameters are detailed in Table 4. The effect of hesperidin consumption on body fat The effect of hesperidin consumption on SBP and was evaluated in 1 study and a significant decrease was DBP was evaluated in 5 studies.11,33,38,39,42 Of these,

856 Nutrition ReviewsVR Vol. 77(12):845–864 3 reported no significant changes11,33,38 and significant subjects after the consumption of 237 mg/day of hesper- decreases were detected in the other 2 studies.39,42 idin in OJ for 12 weeks, compared with basal levels. Rangel Huerta et al39 observed that the consumption of Significant differences were found between both inter- 237 mg/day of hesperidin for 12 weeks reduced the SBP vention groups. and DBP of overweight or obese subjects by 4 mmHg The effect of hesperidin consumption on the and 3 mmHg, respectively, compared with basal levels. QUICKI index was evaluated in 2 studies; neither of No significant differences were observed in a compari- these studies reported any significant changes.11,33 son with the group administered a lower concentration Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 of hesperidin. Morand et al42 reported that the consumption of 292 mg/day of hesperidin – in the form of Results for lipid profile parameters. The characteristics pure hesperidin capsules or provided naturally with 500 of the long-term RCTs included in this review in rela- mL/day of OJ for 4 weeks – reduced the DBP of over- tion to lipid profiles are detailed in Table 5. weight subjects by 5.30 mmHg and 4.50 mmHg, respec- The effect of hesperidin consumption on TC levels tively, compared with basal levels. Significant was evaluated in 8 studies.11,33–35,39–42 Of these, 6 differences were observed in a comparison with the reported no significant changes11,33,34,39,42 and 2 studies control group. reported significant decreases.35,36 Aptekmann and The effect of hesperidin consumption on endothe- Cesar40 found that the TC levels of overweight subjects lial function was evaluated in 3 studies11,33,36: 2 of these were significantly decreased by 0.22 mmol/L, compared studies reported significant increases11,36 and the other with basal levels after 13 weeks of consumption of 54.60 study found no significant changes.33 In subjects with mg/day of hesperetin in OJ. No significant differences metabolic syndrome, Buscemi et al36 observed a signifi- were observed between the intervention group and the cant increase in flow-mediated dilation (FMD) of 2.20% control group. Cesar et al41 reported a significant de- after 1.5 weeks of the consumption of 159.50 mg/day of crease of 0.46 mmol/L in the TC levels of hypercholes- hesperidin in 500 mL/day of OJ. Significant differences terolemic subjects who consumed 42 mg/day of between the intervention group and the control group hesperetin in 750 mL/day of OJ for 8 weeks, compared were observed. Similarly, in subjects with metabolic with the control subjects. No significant differences syndrome, Rizza et al11 reported a significant increase were observed between the intervention and control in FMD of 2.48% after 3 weeks of the consumption of groups. 500 mg/day of hesperidin in capsule form, and the dif- Eight studies evaluated the effect of hesperidin ferences between the intervention and control groups consumption on LDL-c levels.11,33–35,39–42 Of these, 6 were significant. reported no significant changes,11,33–35,39,42 while significant decreases were found in the other 2 studies.40,41 Specifically, compared with the basal level, Results for glucose and insulin levels. The characteristics Aptekmann and Cesar40 observed a significant de- of the long-term RCTs included in this review in crease of 0.44 mmol/L in the LDL-c levels of over- relation to glucose and insulin levels are detailed in weight subjects after 13 weeks of the consumption of Table 4. 54.60mg/dayofhesperetininOJ.Nosignificantdif- The effect of hesperidin consumption on plasma ferences were observed between the intervention and glucose levels was evaluated in 5 studies.11,33,35,39,42 Of control groups. Cesar et al41 observed a significant these, 4 reported no significant changes,11,33,35,42 and a decrease of 0.49 mmol/L in the LDL-c levels of hyper- significant increase was observed in the other study.39 cholesterolemic subjects who consumed 42 mg/day of Specifically, Rangel-Huerta et al39 observed significant hesperetin in OJ 8.5 weeks. No significant differences increases of 0.30 mmol/L and 0.20 mmol/L in the glu- were observed between the intervention and control cose levels of overweight and obese subjects after the groups. consumption of 237 mg/day and 582.50 mg/day of hes- The effect of hesperidin consumption on HDL-c peridin in OJ, respectively, for 12 weeks, compared with levels was evaluated in 8 studies.11,33–35,39–42 No basal levels. Significant differences were observed be- significant changes were detected in 7 of these stud- tween both intervention groups. ies,11,33–35,39,41,42 and the other study reported a sig- Four studies evaluated the effect of hesperidin con- nificant increase.40 In overweight subjects, sumption on plasma insulin levels11,33,39,42: 3 of these Aptekmann and Cesar40 found that the consumption studies reported no significant changes,11,33,42 whereas a of 54.60 mg/day of hesperetin in OJ for 13 weeks in- significant decrease was detected in the other study.39 creased HDL-c levels by 0.23 mmol/L, compared with Rangel-Huerta et al39 noted a significant decrease of basal levels. No significant differences were observed 1.20 mU/mL in the insulin levels of overweight or obese between the intervention and control groups.

Nutrition ReviewsVR Vol. 77(12):845–864 857 858 Table 5 Characteristics of long-term randomized clinical trials included in the systematic review with results for lipid profiles (n ¼ 8) Author, year, Sample Cardiovascular risk Nutritional intervention Results reference size factors Groups Flavanone Product Duration Lipid profile dose dose TC, mmol/L LDL-c, HDL-c, TG, mmol/L APO A, APO B, mmol/L mmol/L mg/dL mg/dL Salden et al n¼6 Overweight or obesity Placebo capsule 0 mg/d 500 mg/d 6 wk NS/NS NS/NS NS/NS NS/NS –– (2016)33 vs (NS) (NS) (NS) (NS) Hesperidin 450 mg/d 500 mg/d capsule Constans et n¼25 Hypercholesterolemia Control drink 0 mg/d 600 mL/d 4wk #0.35/NS #0.32/NS NS/NS NS/NS NS/"5.00 NS/"8.00 al (2015)35 vs (NS) (NS) (NS) (NS) (NS) (NS) Orange juice 213 mg/d 600 mL/d (hesperidin) Rangel- n¼100 Overweight or obesity Orange juice 237 mg/d 500 mL/d 12 wk NS/NS NS/NS NS/NS #0.09/NS NS/#4.00 #4.00/NS Huerta et (hesperidin) (NS) (NS) (NS) (NS) (NS) (NS) al (2015)39 vs Enriched orange 582.5 mg/d 500 mL/d juice (hesperidin) Rizza et al n¼24 Metabolic syndrome Placebo capsule 0 mg/d 1 capsule/d 3 wk NS/NS NS/NS NS/NS NS/NS NS/NS NS/NS (2011)11 vs (P <0.05)* (NS) (NS) (NS) (NS) (NS) Hesperidin 500 mg/d 1 capsule/d capsule Morand et al n¼24 Overweight Placebo capsule 0 mg/d 1 capsule and 4 wk NS/NS/NS NS/NS/NS NS/NS/NS NS/NS/NS –– (2011)42 with control 500 mL/d (NS) (NS) (NS) (NS) drink vs Hesperidin cap- 292 mg/d 1 capsule and sule with con- 500 mL/d trol drink vs Orange juice 292 mg/d 500 mL/d (hesperidin) urto Reviews Nutrition Demonty et n¼194 Overweight Placebo capsule 0 mg/d 4 capsules/d 4 wk NS/NS NS/NS NS/NS NS/NS –– al (2010)34 Hypercholesterolemia vs (NS) (NS) (NS) (NS) Hesperidin 800 mg/d 4 capsules/d capsule Aptekmann n¼26 Overweight No intervention 0 mg/d 0 mL/d 13 wk NS/#0.22 NS/#0.44 NS/#0.23 NS/NS –– V R and Cesar vs (NS) (NS) (NS) (NS) o.77(12):845–864 Vol. (2010)41 Orange juice 54.6 mg/d 500 mL/d (hesperetin) Cesar et al n¼22 Hypercholesterolemia No intervention 0 mg/d 0 mL/d 8.5 wk NS/#0.24 NS/#0.49 NS/NS "0.38/NS –– (2010)41 Vs (NS) (NS) (NS) (NS) Orange juice 42 mg/d 750 mL/d (hesperetin) -, parameter not evaluated; #, significant decrease between basal and final value within each group; ", significant increase between basal and final value within each group; *, significant differences between groups. Abbreviations: APO A, apolipoprotein A; APO B, apolipoprotein B; HDL-c, high density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; NS, no significant differences between

basal and final within each group; (NS), no significant differences between groups; TC, total cholesterol; TG, triglycerides. Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 November 03 on Dent Carly access, site OUP by https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 from Downloaded urto Reviews Nutrition Table 6 Characteristics of long-term randomized clinical trials included in the systematic review with results for coagulation, inflammation, and oxidation biomarkers (n ¼ 6) Author, year, Sample Cardiovascular risk Nutritional intervention Results reference size factors Groups Flavanone Product Duration Coagulation Inflammation biomarkers Oxidation dose dose biomarkers biomarkers V R FIB HOM IL-6, sVCAM-1 sICAM-1 SAA sE-SEL sP-SEL oxLDL, NO o.77(12):845–864 Vol. pg/mL PROTEIN pg/mL Salden et al n¼6 Overweight or obesity Placebo capsule 0 mg/d 500 mg/d 6 wk – – – NS/NS NS/NS – NS/NS NS/NS –– (2016)33 vs (NS) (NS) (NS) (NS) Hesperidin capsule 450 mg/d 500 mg/d Constans et al n¼25 Hypercholesterolemia Control drink 0 mg/d 600 mL/d 4 wk NS/NS NS/NS – NS/NS(NS) NS/NS(NS) – NS/NS(NS) – – – (2015)35 vs (NS) (NS) Orange juice 213 mg/d 600 mL/d (hesperidin) Rangel-Huerta n¼100 Overweight or obesity Orange juice 237 mg/d 500 mL/d 12 wk – – – – – – – – NS/NS – et al (2015)39 (hesperidin) (NS) vs Enriched orange 582.5 mg/d 500 mL/d juice (hesperidin) Buscemi et al n¼31 Metabolic syndrome Control drink 0 mg/d 500 mL/d 1.5 wk – – NS/#3.30 – – – – – – NS/NS (2012)36 vs (P <0.05)* (NS) Orange juice 159.5 mg/d 500 mL/d (hesperidin) Rizza et al n¼24 Metabolic syndrome Placebo capsule 0 mg/d 1 capsule/d 3 wk NS/NS NS/NS – NS/NS NS/NS NS/NS NS/NS ––– (2011)11 vs (NS) (NS) (NS) (NS) (P <0.05)* (NS) Hesperidin capsule 500 mg/d 1 capsule/d Morand et al n¼24 Overweight Placebo capsule 0 mg/d 1 capsule and 4 wk – – NS/NS/NS NS/NS/NS NS/NS/NS – – – – NS/NS/NS (2011)42 with control drink 500 mL/d (NS) (NS) (NS) (NS) vs Hesperidin capsule 292 mg/d 1 capsule and with control drink 500 mL/d vs Orange juice 292 mg/d 500 mL/d (hesperidin) -, parameter not evaluated; #, significant decrease between basal and final value within each group; *, significant differences between groups. Abbreviations: FIB, fibrinogen; HOM, homocysteine; IL-6, interleukin-6; NO, nitric oxide; NS, no significant differences between basal and final within each group; (NS), no significant differences between groups; oxLDL, oxidized low-density lipoprotein; SAA protein, serum amyloid A protein; sE-SEL, soluble E-selectin; sICAM-1, soluble intercellular adhesion molecule-1; sP-SEL, soluble P-selectin; sVCAM-1, soluble vascular cell adhesion molecule-1.

859 Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 November 03 on Dent Carly access, site OUP by https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 from Downloaded Eight studies evaluated the effect of hesperidin con- molecule-1 (sVCAM-1) and soluble intercellular adhe- sumption on TG levels.11,33–35,39–42 Of these, 7 reported sion molecule-1 (s-ICAM-1),11,33,35,42 and the plasma no significant changes11,33–35,39,40,42 and the other study levels of sE-selectin (soluble E-selectin) and sP-selectin reported a significant decrease.39 Compared with basal (soluble P-slectin) were evaluated in 3 studies11,33,35 and levels, Rangel-Huerta et al39 observed a significant de- 1 study,33 respectively. None of these studies detected crease of 0.09 mg/dL in the TG levels of overweight and any significant changes. obese subjects who consumed 237 mg/day of hesperidin The effect of hesperidin consumption on oxidative in OJ for 12 weeks. No significant differences were ob- biomarkers was assessed according to plasma NOx lev- Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 served between the intervention and control groups. els in 2 studies,36,42 but no significant changes were ob- The effects of hesperidin consumption on apolipo- served. Additionally, plasma oxidized low-density protein A-1 (Apo A-1) and apolipoprotein B (Apo B) lipoprotein levels were assessed in 1 study, but no sig- were evaluated in 3 studies,11,35,39 and different results nificant changes were detected.39 were obtained. Specifically, compared with basal levels, Constans et al35 reported a significant increase in Apo Effects of acute hesperidin consumption on A-1 and Apo B levels of 5 mg/dL and 8 mg/dL, respec- cardiovascular risk biomarkers. The effects of acute con- tively, in hypercholesterolemic subjects after the consumption of hesperidin were evaluated according to sumption of 213 mg/day of hesperidin in OJ for 4 vascular parameters (SBP, DBP, and endothelial func- weeks. No significant differences between the intervention) and inflammation biomarkers (sVCAM-1, tion and control groups were observed. Rangel-Huerta s-ICAM-1, sE-selectin, and sP-selectin) in 2 studies, but et al39 noted a significant decrease of 4 mg/dL in the no significant changes in any of the investigated param- Apo A-1 levels and also in the Apo B levels of over- eters were detected. weight or obese subjects who consumed 237 mg/day and 582.50 mg/day of hesperidin in OJ for 12 weeks, DISCUSSION compared with basal levels. No significant differences were observed between the intervention and control The current systematic review presents a summary of groups. In addition, Rizza et al3311 found no significant the available scientific evidence regarding the effects of changes between these two groups. hesperidin consumption on cardiovascular risk biomarkers obtained from animal studies and human RCTs. Results for coagulation, inflammation, and oxidative The results from the animal studies included in the biomarkers. The characteristics of the long-term RCTs present systematic review showed that daily consump- included this review in relation to the biomarkers of co- tion of 50–200 mg/kg of body weight of hesperidin or agulation, inflammation, and oxidation are detailed in hesperetin for a period ranging from 15 days to 24 Table 6. weeks significantly lowered blood glucose levels in type The effect of hesperidin consumption on coagula- 2 diabetic rats and mice. As possible mechanisms of action biomarkers, assessed based on the plasma levels of tion, other experimental studies with rats have sug- fibrinogen and homocysteine, was explored in 2 stud- gested that hesperidin consumption may increase ies,11,35 but neither of these RCTs reported any signifi- hepatic glycolysis and hepatic glucokinase activity and cant changes. decrease hepatic gluconeogenesis and hepatic glucose- In one study, the effect of hesperidin consumption 6-phospatase activity,43 which would inhibit the gluco- on inflammation biomarkers was assessed according to neogenic pathway in liver cells44 and thus prevent the plasma protein serum amyloid A (SAA) levels,11 but no progression of hyperglycemia.43,45 These beneficial significant changes were observed. Inflammation was effects on glucose and insulin levels were not observed also assessed according to plasma IL-6 levels in 2 stud- in the human RCTs included in this systematic review. ies.36,42 Of these, 1 study observed a significant de- However, it is interesting to note that only 5 of the 11 crease,36 but no significant changes were detected in the RCTs included in the review assessed the effects of hes- other study.42 Buscemi et al36 found a significant de- peridin consumption on blood glucose levels, and the crease of 3.30 pg/mL in the IL-6 levels of subjects with population investigated in these RCTs were overweight, metabolic syndrome after the consumption of 159.50 obese, or hypercholesterolemic, whereas the animal mg/day of hesperidin in OJ for 1.5 weeks, compared studies were performed on type 2 diabetic rats. Because with basal levels. Significant differences were observed the types of population investigated in the RCTs that between the intervention and control groups. Four evaluated glucose levels yielded no significant results studies evaluated the effects of hesperidin consumption and because only a few RCTs evaluated the possible ef- on the plasma levels of soluble vascular cell adhesion fect of hesperidin on glucose, more RCTs should be

860 Nutrition ReviewsVR Vol. 77(12):845–864 conducted with type 2 diabetic subjects to assess the OJ for 12–13 weeks, compared with the basal values. effects of hesperidin consumption on glucose and insu- However, both of these studies had some limitations: lin levels in order to confirm the results observed in ani- one was not a placebo-controlled clinical trial,39 and the mals. With respect to insulin levels, no relevant changes other study observed decreases in both the intervention were observed in either the animal studies or the hu- and control groups,40 probably owing to the fact that man RCTs. volunteers tend to pay more attention to their health The animal studies included in the present system- when participating in a study.48 atic review demonstrated that daily consumption of Hesperidin has aroused interest on account of its Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 hesperidin or hesperetin at a dose of 50–200 mg/kg of possible effect on blood pressure because it has been body weight and 1% or 4.60% of total calorie intake suggested that this compound exerts effects similar to improves the lipid profile by significantly reducing those found with other flavonoids, such as quercetin.49 blood levels of TC, LDL-c, and TG in rats and mice In-vitro studies have shown that the improvements in with type 2 diabetes and myocardial ischemia. An in-vi- blood pressure and endothelial dysfunction observed af- tro study showed that the possible mechanism through ter hesperidin consumption may be mediated by a de- which hesperidin improves the lipid profile may involve crease in NADPH oxidase 2, increase in plasma NO the modulation of hepatic lipid metabolism and the in- metabolites, and an inhibitory effect on angiotensin- hibition of Apo B in HepG2.46 In contrast, the results of converting enzyme.50,51 These data suggest that hesperi- the RCTs included in this review did not show the same din may increase the secretion of NO by human endo- conclusive results. In fact, only 240,41 of the 8 articles thelial cells, inhibit cyclic nucleotide phosphodiesterase, that assessed lipid profiles observed a decrease in TC and increase cyclic AMP (adenosine monophosphate) and LDL-c levels. Interestingly, only one study41 and GMP (guanosine monophosphate), thereby exert- assessed the effect of hesperidin on lipid profile in hy- ing a vasorelaxant effect.14,52,53 Nevertheless, according percholesterolemic subjects. This RCT observed marked to the findings of the present review of animal studies decreases of 0.47 mmol/L and 0.49 mmol/L in TC and and RCTs, the consumption of hesperidin has no clear LDL-c levels, respectively,47 after the consumption of 42 effect on DBP and SBP levels. However, it is interesting mg/day of hesperidin in OJ for 8.5 weeks, and this find- to note that the subjects assessed in the included RCTs ing was clinically relevant.47 Thus, hypercholesterolemic were overweight or obese, with no hypertension or ele- subjects constitute an appropriate population for fur- vated blood pressure levels. Therefore, studies that eval- ther evaluation of the specific effects of hesperidin on uate the effect of hesperidin on blood pressure in lipid profile. The differences between the doses of hes- subjects with high blood pressure levels are needed for peridin administered in the animal and human studies us to draw a definitive conclusion about this CVRF. (higher doses were used in the animal studies than in Interestingly, 3 RCTs 11,33,36 included in the present re- the human RCTs) may also have contributed to the dif- view assessed the effects of hesperidin on endothelial ference in the results obtained from these two types of function, and 2 of these11,36 observed improvements in studies. Thus, more human RCTs are needed to better these parameters in subjects with metabolic syndrome understand the effects of hesperidin consumption on and increased CVRFs after 1.5–3 weeks of intervention lipid profile in humans. with 300–500 mg/day of hesperidin in OJ or capsule The present systematic review showed that, in ani- form. Although the available evidence is scarce, it mal models, the consumption of hesperidin does not appears that hesperidin consumption seems likely to in- improve anthropometric parameters, such as body crease endothelial function. Thus, more human RCTs weight and visceral fat. However, it is important to note are needed to determine whether hesperidin decreases that the animal studies included in this review were blood pressure and improves endothelial function in conducted with rats or mice with normal body weight hypertensive or type 2 diabetic populations. and anthropometric parameters for their age; future The results obtained in the present review of RCTs studies should investigate overweight or obese rats or showed that hesperidin has no significant effects on bio- mice to allow more relevant conclusions to be drawn. markers of coagulation, inflammation, and oxidation. Similarly, in the human RCTs, there were no effects of However, few studies have assessed the effect of hesperi- hesperidin on body weight, BMI, and body fat, and only din on these biomarkers in relation to CVDs because al- a limited number of studies have assessed these parame- most all studies have focused on cancer and other ters. Two39,40 of the 3 RCTs that evaluated the effect of chronic diseases.26,54,55 hesperidin consumption on body weight and BMI One factor to consider is the interindividual vari- observed reductions of 1.30–1.80 kg/m2 and 0.30–0.70 ability in hesperidin bioavailability, which may, for ex- kg/m2, respectively, in overweight subjects after daily ample, depend on the microbiota composition of each consumption of 54.60–582.50 mg/day of hesperidin in subject.56,57 Thus, it is possible that different individuals

Nutrition ReviewsVR Vol. 77(12):845–864 861 administered the same dose of hesperidin can absorb not have been considered to a sufficient degree because this compound to different degrees, and therefore, these only 3 RCTs controlled the participants’ diet through vali- individuals would show different effects for the various dated dietary records, and no biomarkers for consump- cardiovascular biomarkers. This could also explain the tion were used in any of the included studies. Therefore, differing results between the studies included in this re- other polyphenol compounds present in the diet may view because none of the studies considered the bio- have been responsible, either partially or entirely, for the availability of hesperidin. observed health effects. In addition, with inadequate mon-

The RCTs included in the present review that ob- itoring of the participants’ diet, it is possible that some Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 served more significant changes39–41 presented many subjects had greater hesperidin intake than others because potential risks of bias, which were classified as unclear they consumed food or beverages with significant risk owing to insufficient information about allocation amounts of hesperidin, potentially affecting the study concealment and blinding of participants, personnel, results of the study. Thus, in nutritional RCTs, monitoring and outcomes, or in terms of including a conflict of in- of the participants’ diet is necessary to avoid confounding terest based on the Cochrane risk of bias criteria. These between other dietary compounds and the dietary inter- unclear risks of bias indicate potential problems related vention. Limiting hesperidin intake as a dietary recom- to the methodological quality of the studies and hence mendation for all participants, monitoring their dietary lead us to question the reliability of the results of the intake, and the use of biomarkers for consumption are RCTs. Therefore, further RCTs are needed with a lower necessary to obtain robust results in this type of study. risk of bias and consequent improvement in quality. Lastly, most of the articles included in this review lacked One strength of this review concerns the standard- statistical data, such as mean differences and their stan- ized methodology that was used. In addition, the in- dard deviation and the standard error or confidence inter- cluded studies were published recently and thus vals for each intervention, as well as their p-values. presented strong scientific evidence, such as RCTs, Consequently, a meta-analysis, which would have pro- along with analyses of their individual risks of bias. vided more conclusive results, as well as a forest plot, Moreover, the novelty of this review lies in the fact that which would have provided a clearer presentation of the it was the first to evaluate the effects of hesperidin con- results, could not be performed. sumption on different CVRFs based on both animal models and human studies. However, the present re- CONCLUSION view has several limitations that warrant discussion. The first is the scarce scientific evidence available from In conclusion, hesperidin consumption was found to human and animal studies that assessed the effects of improve glucose levels and various lipid profile parame- hesperidin on CVRFs. In most studies, the populations ters, such as TC, LDL-c, and TG, in animal models, but used to evaluate the effects of hesperidin on different no definitive conclusion regarding the effects of hesper- CVRFs have not been the most appropriate for reaching idin on different CVRFs in humans can be currently definitive conclusions. Thus, if the objective of a study drawn. Further RCTs of greater quality are needed to is to improve a specific cardiovascular risk factor – for confirm that the results observed in animal models can example, to reduce high serum cholesterol concentrations be translated to the human population and thus to eval- in humans – the recommendation is to include subjects uate whether the administration of hesperidin through that present with symptoms associated with this specific the consumption of citrus food or as a supplement CVRF, such as hypercholesterolemic patients.58 In addi- would serve as a new tool for the prevention and treat- tion, the studies included in this review utilized different ment of CVDs. intervention durations, monitoring approaches, and methods of supplementation. However, the sample size in some of the animal studies was perhaps insufficient for a Acknowledgments robust evaluation of the objectives, and in 2 studies, the doses of hesperidin or hesperetin were not estimated in Author contributions. L.P.-P., R.M.V., and A.P. deter- milligrams, and therefore their dose-dependent effects mined the search strategy. L.P.-P. obtained the studies could not be compared with those of other studies. In ad- identified in database searches for inclusion in the re- dition, dose- and time-dependent effects, as well as the view. L.P.-P. evaluated the quality of the studies, and physiological relevance of the dose used, were not evalu- J.C. and L.C.-P. validated and discussed any discrepan- ated in the animal studies. Also, the possibility of residual cies. The results of the studies were entered by L.P.-P. confounding related to hesperidin bioavailability cannot and validated by L.C.-P. and J.C. L.P.-P. wrote the man- be excluded. Moreover, even though compliance with the uscript, and J.C., L.C.-P., E.L., A.P., R.M.V., and R.S. nutritional intervention is necessary, dietary factors may provided feedback on the manuscript.

862 Nutrition ReviewsVR Vol. 77(12):845–864 Funding. Eurecat, Centre Tecnologic de Catalunya, 20. Potter AS, Foroudi S, Stamatikos A, et al. Drinking carrot juice increases total antioxidant status and decreases lipid peroxidation in adults. Nutr J. Unitat de Nutricio i Salut, Reus, Spain. 2011;10:96. 21. Vargova V, Pytliak M, Mechirova V, et al. Bioavailability in humans of the flavanones hesperidin and narirutin after the ingestion of two doses of orange juice. J Declaration of interest. None of the authors have rele- Agric Food Chem. 2016;4:228–233. vant interests to declare. 22. Selvaraj P, Pugalendi KV. Efficacy of hesperidin on plasma, heart and liver tissue lipids in rats subjected to isoproterenol-induced cardiotoxicity. Exp Toxicol Pathol. 2012;64:449–452. 23. Moher D, Liberati A, Tetzlaff JA. PRISMA 2009 Flow Diagram. Prism Statement 2009;6:e1000097. 24. Iskender H, Dokumacioglu E, Sen TM, et al. The effect of hesperidin and quercetin Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 Supporting Information on oxidative stress, NF-jB and SIRT1 levels in a STZ-induced experimental diabetes model. Biomed Pharmacother. 2017;90:500–508. 25. Dobias L, Petrova M, Vojtko R, et al. Long-term treatment with hesperidin The following Supporting Information is available through improves endothelium-dependent vasodilation in femoral artery of spontane- the online version of this article at the publisher’s website. ously hypertensive rats: the involvement of NO-synthase and Kv channels. Table S1 Characteristics and results of animal Phytother Res. 2016;30:1665–1671. 26. Mahmoud AM, Ashour MB, Abdel-Moneim A, et al. Hesperidin and naringin atten- studies included in the systematic review (n 5 12) uate hyperglycemia-mediated oxidative stress and proinflammatory cytokine pro- Table S2 Characteristics and results of randomized duction in high fat fed/streptozotocin-induced type 2 diabetic rats. J Diabetes Complications. 2012;26:483–490. clinical trials included in the systematic review (n 5 11) 27. Akiyama S, Katsumata S, Suzuki K, et al. Hypoglycemic and hypolipidemic effects of hesperidin and cyclodextrin-clathrated hesperetin in Goto-Kakizaki rats with type 2 diabetes. Biosci Biotechnol Biochem. 2009;73:2779–2782. REFERENCES 28. Jung UJ, Lee MK, Park YB, et al. Effect of citrus flavonoids on lipid metabolism and glucose-regulating enzyme mRNA levels in type-2 diabetic mice. Int J Biochem 1. World Health Organization. Top 10 Causes of Death. 2018. http://www.who.int/ Cell Biol. 2006;38:1134–1145. gho/mortality_burden_disease/causes_death/top_10/en/. Accessed June 5, 2018. 29. Jung UJ, Lee M-K, Jeong K-S, et al. The hypoglycemic effects of hesperidin and 2. Del Rio D, Rodriguez-Mateos A, Spencer JPE, et al. Dietary (poly)phenolics in hu- naringin are partly mediated by hepatic glucose-regulating enzymes in C57BL/ man health: structures, bioavailability, and evidence of protective effects against KsJ-db/db mice. JNutr.2004;134:2499–2503. chronic diseases. Antioxid Redox Signal. 2013;18:1818–1892. 30. Kumar B, Gupta SK, Srinivasan BP, et al. Hesperetin ameliorates hyperglycemia in- 3. Soldati L, Di Renzo L, Jirillo E, et al. The influence of diet on anti-cancer immune duced retinal vasculopathy via anti-angiogenic effects in experimental diabetic responsiveness. J Transl Med. 2018;16:75. rats. Vascul Pharmacol. 2012;57:201–207. 4. Santhakumar AB, Battino M, Alvarez-Suarez JM, Dietary polyphenols: structures, 31. Jia S, Hu Y, Zhang W, et al. Hypoglycemic and hypolipidemic effects of neohesper- bioavailability and protective effects against atherosclerosis.FoodChemToxicol. idin derived from Citrus aurantium L. in diabetic KK-A(y) mice. Food Funct. 2018;113:49–65. 2015;6:878–886. 5. Nabavi SF, Sureda A, Dehpour AR, et al. Regulation of autophagy by polyphenols: 32. Yamamoto M, Jokura H, Hashizume K, et al. Hesperidin metabolite hesperetin-7- paving the road for treatment of neurodegeneration. Biotechnol Adv. O-glucuronide, but not hesperetin-30-O-glucuronide, exerts hypotensive, vasodila- 2017;36:1768–1778. tory, and anti-inflammatory activities. Food Funct. 2013;4:1346. 6. Li C, Schluesener H. Health-promoting effects of the citrus flavanone hesperidin. 33. Salden BN, Troost FJ, de Groot E, et al. Randomized clinical trial on the efficacy of Crit Rev Food Sci Nutr. 2017;57:613–631. hesperidin 2S on validated cardiovascular biomarkers in healthy overweight indi- 7. USDA Economic Research Service. Food Availability (Per Capita) Data System.535 viduals. Am J Clin Nutr. 2016;104:1523–1533. Economic Research Service, Department of Agriculture. 2018. https://data.nal. 34. Demonty I, Lin Y, Zebregs Y, et al. The citrus flavonoids hesperidin and naringin usda.gov/dataset/food-availability-capita-data-system. Accessed June 5, 2018. do not affect serum cholesterol in moderately hypercholesterolemic men and 8. Vogiatzoglou A, Mulligan AA, Lentjes MAH, et al. Flavonoid intake in European women. JNutr. 2010;140:1615–1620. adults (18 to 64 years). PLoS One 2015;10:e0128132. 35. Constans J, Bennetau-Pelissero C, Martin JF, et al. Marked antioxidant effect 9. Rafiq S, Kaul R, Sofi SA, et al. Citrus peel as a source of functional ingredient: a re- of orange juice intake and its phytomicronutrients in a preliminary random- view. J Saudi Soc Agric Sci. 2018;17:351–358. ized cross-over trial on mild hypercholesterolemic men. Clin Nutr. 10. Bredsdorff L, Nielsen ILF, Rasmussen SE, et al. Absorption, conjugation and excre- 2015;34:1093–1100. tion of the flavanones, naringenin and hesperetin from alpha-rhamnosidase- 36. Buscemi S, Rosafio G, Arcoleo G, et al. Effects of red orange juice intake on endo- treated orange juice in human subjects. Br J Nutr. 2010;103:1602–1609. thelial function and inflammatory markers in adult subjects with increased cardio- 11. Rizza S, Muniyappa R, Iantorno M, et al. Citrus polyphenol hesperidin stimulates vascular risk. Am J Clin Nutr. 2012;95:1089–1095. production of nitric oxide in endothelial cells while improving endothelial func- 37. Schar MY, Curtis PJ, Hazim S, et al. Orange juice-derived flavanone and phenolic tion and reducing inflammatory markers in patients with metabolic syndrome. J metabolites do not acutely affect cardiovascular risk biomarkers: a randomized, Clin Endocrinol Metab. 2011;96:782–792. placebo-controlled, crossover trial in men at moderate risk of cardiovascular dis- 12. Chiou CS, Lin JW, Kao PF, et al. Effects of hesperidin on cyclic strain-induced endo- ease. Am J Clin Nutr. 2015;101:931–938. thelin-1 release in human umbilical vein endothelial cells. Clin Exp Pharmacol 38. Kean RJ, Lamport DJ, Dodd GF, et al. Chronic consumption of flavanone-rich or- Physiol. 2008;35:938–943. ange juice is associated with cognitive benefits: an 8-wk, randomized, double- 13. Jin YR, Han XH, Zhang YH, et al. Antiplatelet activity of hesperetin, a bioflavonoid, blind, placebo-controlled trial in healthy older adults. Am J Clin Nutr. is mainly mediated by inhibition of PLC-c2 phosphorylation and cyclooxygenase- 2015;101:506–514. 1 activity. Atherosclerosis. 2007;194:144–152. 39. Rangel-Huerta OD, Aguilera CM, Martin MV, et al. Normal or high polyphenol con- 14. Ikemura M, Sasaki Y, Giddings JC, et al. Preventive effects of hesperidin, glucosyl centration in orange juice affects antioxidant activity, blood pressure, and body hesperidin and naringin on hypertension and cerebral thrombosis in stroke-prone weight in obese or overweight adults. JNutr.2015;145:1808–1816. spontaneously hypertensive rats. Phytother Res. 2012;26:1272–1277. 40. Aptekmann NP, Cesar TB. Orange juice improved lipid profile and blood lactate of 15. Ferreira PS, Spolidorio LC, Manthey JA, et al. Citrus flavanones prevent systemic in- overweight middle-aged women subjected to aerobic training. Maturitas. flammation and ameliorate oxidative stress in C57BL/6J mice fed high-fat diet. 2010;67:343–347. Food Funct. 2016;7:2675–2681. 41. Cesar TB, Aptekmann NP, Araujo MP, et al. Orange juice decreases low-density li- 16. Wang X, Hasegawa J, Kitamura Y, et al. Effects of hesperidin on the progression of poprotein cholesterol in hypercholesterolemic subjects and improves lipid trans- hypercholesterolemia and fatty liver induced by high-cholesterol diet in rats. J fer to high-density lipoprotein in normal and hypercholesterolemic subjects. Nutr Pharmacol Sci. 2011;117:129–138. Res. 2010;30:689–694. 17. Johnsen SP, Overvad K, Stripp C, et al. Intake of fruit and vegetables and the risk 42. Morand C, Dubray C, Milenkovic D, et al. Hesperidin contributes to the vascular of ischemic stroke in a cohort of Danish men and women. Am J Clin Nutr. protective effects of orange juice: a randomized crossover study in healthy volun- 2003;78:57–64. teers. Am J Clin Nutr. 2011;93:73–80. 18. Dauchet L, Ferrie`res J, Arveiler D, et al. Frequency of fruit and vegetable consump- 43. Shen W, Xu Y, Lu YH. Inhibitory effects of Citrus flavonoids on starch digestion tion and coronary heart disease in France and Northern Ireland: the PRIME study. and antihyperglycemic effects in HepG2 cells. J Agric Food Chem. Br J Nutr. 2004;92:963–972. 2012;60:9609–9619. 19. Asgary S, Keshvari M. Effects of Citrus sinensis juice on blood pressure. ARYA 44. Constantin RP, Constantin RP, Bracht A, et al. Molecular mechanisms of citrus fla- Atheroscler. 2013;9:98–101. vanones on hepatic gluconeogenesis. Fitoterapia. 2014;92:148–162.

Nutrition ReviewsVR Vol. 77(12):845–864 863 45. Jayaraman R, Subramani S, Sheik Abdullah SH, et al. Antihyperglycemic effect of 52. Liu L, Xu D-M, Cheng Y-Y. Distinct effects of naringenin and hesperetin on nitric hesperetin, a citrus flavonoid, extenuates hyperglycemia and exploring the poten- oxide production from endothelial cells. J Agric Food Chem. 2008;56:824–829. tial role in antioxidant and antihyperlipidemic in streptozotocin-induced diabetic 53. Orallo F, Alvarez E, Basaran H, et al. Comparative study of the vasorelaxant activity, rats. Biomed Pharmacother. 2018;97:98–106. superoxide-scavenging ability and cyclic nucleotide phosphodiesterase-inhibitory 46. Lin Y, Vermeer MA, Bos W, et al. Molecular structures of citrus flavonoids deter- effects of hesperetin and hesperidin. Naunyn Schmiedebergs Arch Pharmacol. mine their effects on lipid metabolism in HepG2 cells by primarily suppressing 2004;370:452–463. apoB secretion. J Agric Food Chem. 2011;59:4496–4503. 54. Adefegha SA, Rosa Leal DB, Olabiyi AA, et al. Hesperidin attenuates inflammation 47. Catapano AL, Graham I, De Backer G, et al. 2016 ESC/EAS guidelines for the man- and oxidative damage in pleural exudates and liver of rat model of pleurisy. agement of dyslipidaemias. Eur Heart J. 2016;37:2999–3058. Redox Rep. 2017;22:563–571. 48. McKee M, Britton A, Black N, et al. Methods in health services research. 55. Carballo-Villalobos AI, Gonzalez-Trujano ME, Alvarado-Vazquez N, et al. Pro-in- Interpreting the evidence: choosing between randomised and non-randomised flammatory cytokines involvement in the hesperidin antihyperalgesic effects at studies. BMJ. 1999;319:312–315. peripheral and central levels in a neuropathic pain model. Downloaded from https://academic.oup.com/nutritionreviews/article/77/12/845/5528274 by OUP site access, Carly Dent on 03 November 2020 49. Bru¨ll V, Burak C, Stoffel-Wagner B, et al. Effects of a quercetin-rich onion skin Inflammopharmacology. 2017;25:259–265. extract on 24 h ambulatory blood pressure and endothelial function in 56. Andersson A, N€alsen C, Tengblad S, et al. Fatty acid composition of skeletal overweight-to-obese patients with (pre-)hypertension: a randomised double- muscle reflects dietary fat composition in humans. Am J Clin Nutr. blinded placebo-controlled cross-over trial. Br J Nutr. 2015;114:1263–1277. 2002;76:1222–1229. 50. Actis-Goretta L, Ottaviani JI, Fraga GC. Inhibition of angiotensin converting en- 57. Vallejo F, Larrosa M, Escudero E, et al. Concentration and solubility of flavanones zyme activity by flavanol-rich foods. J Agric Food Chem. 2006;54:229–234. in orange beverages affect their bioavailability in humans. J Agric Food Chem. 51. Wunpathe C, Potue P, Maneesai P, et al. Hesperidin suppresses renin-angiotensin 2010;58:6516–6524. system mediated NOX2 over-expression and sympathoexcitation in 2K-1C hyper- 58. Brown L, Rosner B, Willett WW, et al. Cholesterol-lowering effects of dietary fiber: tensive rats. Am J Chin Med. 2018;46:1–17. a meta-analysis. Am J Clin Nutr. 1999;69:30–42.

864 Nutrition ReviewsVR Vol. 77(12):845–864 Supporting Information 1. Table S1. Characteristics and results of animal studies included in the systematic review (N=12). AUTHOR, STUDY EXPERIMENTAL INTERVENTION DOSE OF THE DURATION OF BIOMARKERS RESULTS YEAR, DESIGN ANIMAL GROUPS (N) FLAVANONE AND INTERVENTION REFERENCE ROUTE OF ADMINISTRATION Iskender H et Controlled 4-6-week old male Control group (n=10) vs 0 mg vs 100 mg/kg 15 days Body weight, -Body weight (g): al. (2017) S1 Wistar albino rats intervention group (n=10) body weight/day of glucose Control group = B: 235.80 ± 14.35; F: 190.51 ± 10.16. NDA treated with STZ hesperidin in aqueous Intervention group = B: 242.50 ± 8.36; F: 205.54 ± 11.18. NDA suspension orally (type 2 diabetes) No significant differences between intervention group and control group at the end of the study.

-Glucose (mmol/L): Control group = B: NDA; F: 33.45 ± 1.12. NDA Intervention group = B: NDA; F: 24.20 ± 0.79. NDA Significant differences between intervention group and control group at the end of the study. p<0.05 Dobias L et al. Randomized, 15-week old Control group (n=13) vs 0 mg (corresponding 4 weeks Body weight, SBP -Body weight (g): (2016) S2 controlled, spontaneously intervention group (n=13) volume of distilled Control group = B: 282.00 ± 2.50; F: 301.00 ± 2.40. NDA not blinded hypertensive male water) vs 50 mg/kg Intervention group = B: 290.00 ± 2.70; F: 313.00 ± 2.60. NDA rats body weight/day of No significant differences between intervention group and control group at the end of the study. hesperidin orally -SBP (mm Hg): suspended in distilled Control group = B: 169.00 ± 2.20; F: 168.00 ± 2.62. NDA water Intervention group = B: 161.00 ± 1.70; F: 166.00 ± 2.12. NDA No significant differences between intervention group and control group at the end of the study.

Ferreira PS et Randomized, 9-week old male Control group (n=10) vs 0 mg vs 100 mg/kg 4 weeks Weight gain, -Weight gain (g): al. (2016) S3 controlled C57BL/6J mice with intervention group (n=10) body weight/day of visceral fat, Control group = B: NDA; F: 7.20 ± 3.60. NDA systemic hesperidin orally added glucose, TC, Intervention group = B: NDA; F: 5.60 ± 0.90. NDA inflammation caused to the regular diet HDL-c, LDL-c, No significant differences between intervention group and control group at the end of the study. by high fat diet TG, IL-6 -Visceral fat (%):

Control group = B: NDA; F: 4.50 ± 1.90. NDA Intervention group = B: NDA; F: 3.80 ± 1.40. NDA No significant differences between intervention group and control group at the end of the study.

-Glucose (mmol/L): Control group = B: NDA; F: 20.81 ± 3.27. NDA Intervention group = B: NDA; F: 18.70 ± 4.72. NDA No significant differences between intervention group and control group at the end of the study.

-TC (mmol/L): Control group = B: NDA; F: 3.59 ± 0.05. NDA Intervention group = B: NDA; F: 3.17 ± 0.52. NDA No significant differences between intervention group and control group at the end of the study.

-HDL-c (mmol/L): Control group = B: NDA; F: 2.16 ± 0.03. NDA Intervention group = B: NDA; F: 2.00 ± 0.39. NDA No significant differences between intervention group and control group at the end of the study.

-LDL-c (mmol/L): Control group = B: NDA; F: 1.04 ± 0.31. NDA Intervention group = B: NDA; F: 0.75 ± 0.23. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

-TG (mmol/L): Control group = B: NDA; F: 1.85 ± 0.16. NDA Intervention group = B: NDA; F: 2.18 ± 0.23. NDA No significant differences between intervention group and control group at the end of the study.

-IL-6 (pg/mL): Control group = B: NDA; F: 64.40 ± 44.30. NDA Intervention group = B: NDA; F: 5.76 ± 4.32. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

Jia S et al. Randomized, 8-weeks old male Control group (n=10) vs 0 mg (water) vs 50 6 weeks Body weight, -Body weight (g): (2015) S4 controlled KK-Ay mice (type 2 Intervention group (n=10) mg/kg body weight/day glucose, TC, TG Control group = B: 38.39 ± 1.12; F: 43.18 ± 0.80. NDA diabetic) of neohesperidin Intervention group = B: 38.52 ± 1.17; F: 42.16 ± 1.73. NDA (derived from No significant differences between intervention group and control group at the end of the study. hesperidin) by gavage -Glucose (mmol/L): Control group = B: 10.16 ± 0.44; F: 20.91 ± 1.86. NDA Interventional group = B: 10.28 ± 0.35; F: 13.18 ± 2.72. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

-TC (mmol/L): Control group = B: NDA; F: 6.10 ± 0.26. NDA Intervention group = B: NDA; F: 4.60 ± 0.36. NDA Significant differences between intervention group and control group at the end of the study. p<0.01

-TG (mmol/L): Control group = B: NDA; F: 3.73 ± 0.21. NDA Intervention group = B: NDA; F: 1.68 ± 0.22. NDA Significant differences between intervention group and control group at the end of the study. p<0.001

Yamamoto M. Controlled 14-18-weeks old Control group (n=4) vs 0 mg vs 5 mg/kg body 3 minutes (acute SBP, DBP -SBP (mm Hg): (2013) S5 male hypertensive Hesperetin supplement weight/day of each study) Control group = B: 193.40 ± 4.20; F: -3.50 ± 0.40. NDA rats (n=4) vs Hesperetin-7-O- flavanone by Hesperetin supplement = B: 194.10 ± 4.70; F: -9.90 ± 1.70. NDA β-D-glucuronide intravenous H7 supplement = B: 197.00 ± 3.40; F: -8.70 ± 0.80. NDA supplement (H7 administration H3 supplement = B: 201.30 ± 4.20; F: -4.10 ± 0.80. NDA supplement) (n=4) vs Significant differences between hesperetin supplement and control group at the end of the study. p<0.01 Hesperetin-3’-O-β-D- Significant differences between H7 supplement and control group at the end of the study. p<0.05 glucuronide supplement -DBP (mm Hg): (H3 supplement) (n=4) Control group = B: 174.90 ± 3.80; F: data not shown. NDA Hesperetin supplement = B: 173.80 ± 3.90; F: data not shown. NDA H7 supplement = B: 179.80 ± 2.80; F: data not shown. NDA H3 supplement = B: 186.90 ± 6.00; F: data not shown. NDA No significant differences between the 4 groups at the end of the study.

Kumar B et al. Controlled STZ treated Wistar Control group (n=16) vs 0 mg vs 200 mg/kg 24 weeks Weight gain, -Weight gain (%): (2012) S6 albino rats (type 2 intervention group (n=16) body weight/day of glucose Control group = F: 22.34%. NDA diabetes) hesperetin by gavage Intervention group = F: 45.35%. NDA NDA about differences between interventional group and control group.

-Glucose (mmol/L): Control group = B: NDA; F: 27.67 ± 1.03. NDA Intervention group = B: NDA; F: 21.68 ± 2.06. NDA Significant differences between intervention group and control group at the end of the study. p<0.001

Mahmoud AM Controlled STZ treated while Control group (n=6) vs 0 mg vs 50 mg/kg body 30 days Glucose, insulin, -Glucose (mmol/L): et al. (2012) S7 male albino rats intervention group (n=6) weight/day of NO Control group = B: NDA; F: 16.37 ± 0.23. NDA (type 2 diabetes) hesperidin in aqueous Intervention group = B: NDA; F: 6088 ± 0.22. NDA suspension orally Significant differences between intervention group and control group at the end of the study. p<0.01

-Insulin (µU/mL): Control group = B: NDA; F: 15.50 ± 0.76. NDA Intervention group = B: NDA; F: 21.55 ± 1.13. NDA Significant differences between intervention group and control group at the end of the study. p<0.01

-NO (µmol/L): Control group = B: NDA; 13.60 ± 0.53. NDA Intervention group = B: NDA; F: 8.52 ± 0.48. NDA Significant differences between intervention group and control group at the end of the study. p<0.01

Selvaraj P et al. Randomized, Isoproterenol Control group (n=6) vs 0 mg vs 200 mg/kg 7 days TC, HDL-c, LDL- -TC (mmol/L): (2012) S8 controlled hydrochloride intervention group (n=6) body weight/day of c, TG Control group = B: NDA; F: 2.72 ± 0.26. NDA treated male albino hesperidin dissolved in Intervention group = B: NDA; F: 2.32 ± 0.18. NDA Wistar rats carboxyl methyl- Significant differences between intervention group and control group at the end of the study. p<0.05 (myocardial cellulose post-orally -HDL-c (mmol/L): ischemia) Control group = B: NDA; F: 0.79 ± 0.06. NDA Intervention group = B: NDA; F: 1.13 ± 0.11. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

-LDL-c (mmol/L): Control group = B: NDA; F: 1.54 ± 0.13. NDA Intervention group = B: NDA; F: 0.87 ± 0.05. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

-TG (mmol/L): Control group = B: NDA; F: 0.87 ± 0.07. NDA Intervention group = B: NDA; F: 0.69 ± 0.04. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

Wang X et al. Controlled 8-weeks old male Control group (n=15) vs 0 mg vs 0.08 % 12 weeks Body weight, -Body weight (g): (2011) S9 Wistar rats treated intervention group (n=15) TCD/day of hesperidin SBP, DBP Control group = B: NDA; F: 544.50 ± 4.90. NDA with high- Intervention group = B: NDA; F: 542.00 ± 9.00. NDA cholesterol diet (2% No significant differences between interventional group and control group at the end of the study. TCD/day) -SBP (mm Hg): Control group = NDA. Intervention group = NDA. No significant differences between interventional group and control group at the end of the study.

-DBP (mm Hg): Control group = NDA. Intervention group = NDA. No significant differences between interventional group and control group at the end of the study.

Akiyama S et Randomized 3-week old male GK Control group (n=6) vs 0% TCD/day vs 1% 4 weeks Body weight, -Body weight (g): al. (2009) S10 rats (type 2 diabetes) intervention group 1 TCD/day of hesperidin glucose, insulin, Control group = B: 53.67 ± 1.48; F: 179.97 ± 1.82. NDA (n=6) vs intervention vs 4.6% TCD/day of TC, TG Intervention group 1 = B: 52.98 ± 1.45; F: 172.30 ± 5.83. NDA group 2 (n=6) hesperidin Intervention group 2 = B: 52.81 ± 1.57; F: 166.41 ± 2.36. NDA Significant differences between intervention group 2 and control group + intervention group 1 at the end of the study. p<0.05

-Glucose (mmol/L): Control group = B: 6.09 ± 0.19; F: 7.25 ± 0.35. NDA Intervention group 1 = B: 6.19 ± 0.18; F: 5.78 ± 0.22. NDA Intervention group 2 = B: 6.01 ± 0.14; F: 5.64 ± 0.17. NDA Significant differences between intervention group 2 and control group at the end of the study. p<0.05

-Insulin (µU/mL): Control group = B: NDA; F: 116.17 ± 55.53. NDA Intervention group 1 = B: NDA; F: 14.04 ± 9.36. NDA Intervention group 2 = B: NDA; F: 25.53 ± 21.91. NDA Significant differences between intervention group 2 and control group at the end of the study. p<0.05

-TC (mmol/L): Control group = B: NDA; F: 4.76 ± 0.14. NDA Intervention group 1 = B: NDA; F: 3.05 ± 0.11. NDA Intervention group 2 = B: NDA; F: 2.25 ± 0.10. NDA Significant differences between intervention groups and control group at the end of the study. p<0.05

-TG (mmol/L): Control group = B: NDA; F: 1.30 ± 0.16. NDA Intervention group 1 = B: NDA; F: 0.64 ± 0.06. NDA Intervention group 2 = B: NDA; F: 0.39 ± 0.03. NDA Significant differences between intervention groups and control group at the end of the study. p<0.05

Jung UJ et al. Controlled 5-weeks old male Control group (n=10) vs 0 mg vs 0.20 g/kg body 5 weeks Glucose, TC, -Glucose (mmol/L): (2006) S11 C57BL/KsJ-db/db intervention group (n=10) weight/day of HDL-c Control group = B: 21.06 ± 1.05; F: 39.66 ± 1.83. NDA mice (type 2 hesperidin added to the Intervention group = B: 20.66 ± 1.07; F: 31.82 ± 1.11. NDA diabetes) regular diet Significant differences between intervention group and control group at the end of the study. p<0,05

-TC (mmol/L): Control group = B: NDA; F: 5.62 ± 0.12. NDA Intervention group = B: NDA; F: 4.81 ± 0.19. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

-HDL-c (mmol/L): Control group = B: NDA; F: 1.06 ± 0.07. NDA Intervention group = B: NDA; F: 1.17 ± 0.07. NDA No significant differences between intervention group and control group at the end of the study.

-TG (mmol/L): Control group = B: NDA; F: 3.32 ± 0.19. NDA Intervention group = B: NDA; F: 1.58 ± 0.22. NDA Significant differences between intervention group and control group at the end of the study. p<0.05

Jung UJ et al. Controlled 5-weeks old male Control group (n=10) vs 0 mg vs 0.20 g/kg body 5 weeks Insulin -Insulin (µU/mL): (2004) S12 C57BL/KsJ-db/db intervention group (n=10) weight/day of Control group = B: NDA, F: 29.10 ± 1.81. NDA mice (type 2 hesperidin added to the Intervention group = B: NDA; F: 47.18 ± 0.59. NDA diabetes) regular diet Significant differences between intervention group and control group at the end of the study. p<0.05

STZ, streptozotocin; NDA, no data available about p-valor between basal and final values within each group; SBP, systolic blood pressure; TC, total cholesterol; HDL-c, high density lipoprotein cholesterol; LDL-c, low density lipoprotein cholesterol; TG, triglycerides; IL-6, interleukin-6; DBP, diastolic blood pressure; NO, nitric oxide; TCD, total calorie diet.

REFERENCES

S1. Iskender H, Dokumacioglu E, Sen TM, Ince I, Kanbay Y, Saral S. The effect of hesperidin and quercetin on oxidative stress, NF-κB and SIRT1 levels in a STZ-induced experimental diabetes model. Biomed Pharmacother. 2017;90:500-508.

S2. Dobiaš L, Petrová M, Vojtko R, Kristová V. Long-term Treatment with Hesperidin Improves Endothelium-dependent Vasodilation in Femoral Artery of Spontaneously Hypertensive Rats:

The Involvement of NO-synthase and K v Channels. Phyther Res. 2016;30(10):1665-1671.

S3. Ferreira PS, Spolidorio LC, Manthey JA, et al. Citrus flavanones prevent systemic inflammation and ameliorate oxidative stress in C57BL/6J mice fed high-fat diet. Food Funct.

2016;7(6):2675-2681.

S4. Jia S, Hu Y, Zhang W, et al. Hypoglycemic and hypolipidemic effects of neohesperidin derived from Citrus aurantium L. in diabetic KK-A y mice. Food Funct. 2015;6(3):878-886.

S5. Yamamoto M, Jokura H, Hashizume K, et al. Hesperidin metabolite hesperetin-7-O-glucuronide, but not hesperetin-3′-O-glucuronide, exerts hypotensive, vasodilatory, and anti inflammatory activities. Food Funct. 2013;4(9):1346.

S6. Kumar B, Gupta SK, Srinivasan BP, Nag TC, Srivastava S, Saxena R. Hesperetin ameliorates hyperglycemia induced retinal vasculopathy via anti-angiogenic effects in experimental diabetic rats. Vascul Pharmacol. 2012;57(5-6):201-207.

S7. Mahmoud AM, Ashour MB, Abdel-Moneim A, Ahmed OM. Hesperidin and naringin attenuate hyperglycemia-mediated oxidative stress and proinflammatory cytokine production in high fat fed/streptozotocin-induced type 2 diabetic rats. J Diabetes Complications. 2012;26(6):483-490.

S8. Selvaraj P, Pugalendi KV. Efficacy of hesperidin on plasma, heart and liver tissue lipids in rats subjected to isoproterenol-induced cardiotoxicity. Exp Toxicol Pathol. 2012;64(5):449-452.

S9. Wang X, Hasegawa J, Kitamura Y, et al. Effects of Hesperidin on the Progression of Hypercholesterolemia and Fatty Liver Induced by High-Cholesterol Diet in Rats. J Pharmacol Sci.

2011;117(3):129-38.

S10. Akiyama S, Katsumata S, Suzuki K, Nakaya Y, Ishimi Y, Uehara M. Hypoglycemic and hypolipidemic effects of hesperidin and cyclodextrin-clathrated hesperetin in Goto-Kakizaki rats with type 2 diabetes. Biosci Biotechnol Biochem. 2009;73(12):2779-2782.

S11. Jung UJ, Lee MK, Park YB, Kang MA, Choi MS. Effect of citrus flavonoids on lipid metabolism and glucose-regulating enzyme mRNA levels in type-2 diabetic mice. Int J Biochem Cell

Biol. 2006;38(7):1134-45.

S12. Jung UJ, Lee M-K, Jeong K-S, Choi M-S. The hypoglycemic effects of hesperidin and naringin are partly mediated by hepatic glucose-regulating enzymes in C57BL/KsJ-db/db mice. J

Nutr. 2004;134(10):2499-2503.

Supporting Information 2. Table 2S. Characteristics and results of randomized clinical trials included in the systematic review (N=11). AUTHOR, STUDY POPULATION SAMPLE SIZE, NUTRITIONAL DOSE OF THE DURATION OF METHOD TO BIOMARKERS RESULTS YEAR, DESIGN POPULATION INTERVENTION FLAVANONE INTERVENTION CHECK REFERENCE AGE AND INTERVENTION HEALTH COMPLIANCE STATUS Salden BN et al. Randomized, European 65 subjects 500 mg/day of 0 mg/day in 6 weeks - SBP, DBP, FMD, CHRONIC FOLLOW-UP (2016)S1 placebo- (Holland) 34-69 years placebo capsule placebo capsule vs glucose, insulin, VASCULAR PARAMETERS: controlled, CVRF: (n=32) vs 500 450 mg/day in QUICKI, TC, -SBP (mm Hg): double-blind, overweight or mg/day of hesperidin hesperidin capsule LDL-c, HDL-c, Placebo capsule = B: 131.00 ± 3.00; F: 129.00 ± 2.00. NS parallel, chronic obesity capsule (n=33) TG, sVCAM-1, Hesperidin capsule = B: 135.00 ± 2.00; F: 130.00 ± 2.00. NS and acute sICAM-1, sE- No significant differences between placebo capsule and hesperidin capsule at the end selectin, sP- of the study. selectin -DBP (mm Hg): Placebo capsule = B: 80.00 ± 2.00; F: 81.00 ± 2.00. NS Hesperidin capsule = B: 83.00 ± 1.00; F: 81.00 ± 2.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-FMD (%): Placebo capsule = B: 5.57 ± 0.51; F: 5.43 ± 0.47. NS Hesperidin capsule = B: 4.50 ± 0.51; F: 4.29 ± 0.47. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

GLUCOSE METABOLISM: -Glucose (mmol/L): Placebo capsule = B: 5.00 ± 0.10; F: 5.00 ± 0.10. NS Hesperidin capsule = B: 4.90 ± 0.10; F: 5.00 ± 0.10. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-Insulin (µU/mL): Placebo capsule = B: 8.35 ± 3.02; F: 7.63 ± 2.88. NS Hesperidin capsule = B: 10.66 ± 3.02; F: 11.66 ± 3.02. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-QUICKI: Placebo capsule = B: 0.42 ± 0.01; F: 0.23 ± 0.01. NS Hesperidin capsule = B: 0.34 ± 0.01; F: 0.42 ± 0.01. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

LIPID PROFILE: -TC (mmol/L): Placebo capsule = B: 5.70 ± 0.20; F: 5.60 ± 0.20. NS Hesperidin capsule = B: 5.50 ± 0.20; F: 5.40 ± 0.20. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-LDL-c (mmol/L): Placebo capsule = B: 3.60 ± 0.20; F: 3.60 ± 2.00. NS Hesperidin supplement = B: 3.50 ± 0.20; F: 3.40 ± 0.20. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-HDL-c (mmol/L): Placebo capsule = B: 1.50 ± 0.50; F: 1.50 ± 0.10. NS Hesperidin capsule = B: 1.50 ± 0.10; F: 1.50 ± 0.10. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-TG (mmol/L): Placebo capsule = B: 1.30 ± 0.10; F: 1.30 ± 0.10. NS Hesperidin capsule = B: 1.30 ± 0.10; F: 1.30 ± 0.10. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

INFLAMMATION BIOMARKERS: -sVCAM-1 (ng/mL): Placebo capsule = B: 214.00 ± 10.00; F: 215.00 ± 10.00. NS Hesperidin capsule = B: 210.00 ± 10.00; F: 190.00 ± 10.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sICAM-1 (ng/mL): Placebo capsule = B: 107.00 ± 5.00; F: 107.00 ± 5.00. NS Hesperidin capsule = B: 110.00 ± 5.00; F: 100.00 ± 5.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sE-selectin (ng/mL): Placebo capsule = B: 12.00 ± 1.00; F: 11.00 ± 2.00. NS Hesperidin capsule = B: 11.00 ± 1.00; F: 12.00 ± 2.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sP-selectin (ng/mL): Placebo capsule = B: 78.00 ± 5.00; F: 83.00 ± 5.00. NS Hesperidin capsule = B: 94.00 ± 5.00; F: 83.00 ± 5.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

ACUTE FOLLOW-UP VASCULAR PARAMETERS: -FMD (%) at baseline: Placebo capsule = B: 5.57 ± 0.51; T2h: 5.08 ± 0.53. NS Hesperidin capsule = B: 4.50 ± 0.51; T2h: 4.23 ± 0.51. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-FMD (%) after 6 weeks of supplementation: Placebo capsule = B: 5.57 ± 0.51; T2h: 5.08 ± 0.53. NS Hesperidin capsule = B: 4.21 ± 0.48; T2h: 4.38 ± 0.51. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

INFLAMMATION BIOMARKERS: -Acute sVCAM-1 (ng/mL) at baseline: Placebo capsule = B: 214.10 ± 10.00; T2h: 213.00 ± 9.00; T4h: 206.00 ± 10.00. NS Hesperidin capsule = B: 210.00 ± 10.00; T2h: 207.00 ± 9.00; T4h: 209.00 ± 10.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sVCAM-1 (ng/mL) after 6 weeks of supplementation: Placebo capsule = B: 215.00 ± 10.00; T2h: 208.00 ± 9.00; T4h: 200.00 ± 10.00. NS Hesperidin capsule = B: 190.00 ± 10.00; T2h: 185.00 ± 9.00; T4h: 180.00 ± 10.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sICAM-1 (ng/mL) at baseline: Placebo capsule = B: 107.00 ± 5.00; T2h: 105.00 ± 4.00; T4h: 109.00 ± 5.00. NS Hesperidin capsule = B: 110.00 ± 5.00; T2h: 100.00 ± 4.00; T4h: 104.00 ± 5.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sICAM-1 (ng/mL) after 6 weeks of supplementation: Placebo capsule = B: 107.00 ± 5.00; T2h: 105.00 ± 5.00; T4h: 102.00 ± 5.00. NS Hesperidin capsule = B: 100.00 ± 5.00; T2h: 98.00 ± 5.00; T4h: 96.00 ± 5.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sE-selectin (ng/mL) at baseline: Placebo capsule = B: 12.00 ± 1.00; T2h: 12.00 ± 2.00; T4h: 12.00 ± 2.00. NS Hesperidin capsule = B: 11.00 ± 1.00; T2h: 10.00 ± 2.00; T4h: 11.00 ± 2.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sE-selectin (ng/mL) after 6 weeks of supplementation: Placebo capsule = B: 11.00 ± 2.00; T2h: 9.00 ± 2.00; T4h: 10.00 ± 2.00. NS Hesperidin capsule = B: 12.00 ± 2.00; T2h: 11.00 ± 2.00; T4h: 11.00 ± 2.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sP-selectin (ng/mL) at baseline: Placebo capsule = B: 78.00 ± 5.00; T2h: NDA; T4h: 83.00 ± 4.00. NS Hesperidin capsule = B: 94.00 ± 5.00; T2h: NDA; T4h: 90.00 ± 4.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sP-selectin (ng/mL) after 6 weeks of supplementation: Placebo capsule = B: 83.00 ± 5.00; T2h: NDA; T4h: 94.00 ± 5.00. NS Hesperidin capsule = B: 83.00 ± 5.00; T2h: NDA; T4h: 89.00 ± 5.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

Kean RJ et al. Randomized, European (United 37 subjects 500 mL/day of low 64 mg/day of 8 weeks SBP, DBP VASCULAR PARAMETERS: (2015)S2 double-blind, Kingdom) 61-70 years flavanone drink vs hesperidin vs 549 -SBP (mm Hg): crossover, CVRF: 500 mL/day of high mg/day of Low flavanone drink on women = B: 126.00 ± 2.70; F: 125.00 ± 4.00. NS chronic overweight flavanone drink hesperidin High flavanone drink on women = B: 123.00 ± 3.20; F: ± 124.00 ± 2.40. NS Low flavanone drink on men = B: 132.00 ± 4.30; F: 136.00 ± 4.00. NS High flavanone drink on men = B: 135.00 ± 5.60; F: 133.00 ± 3.40. NS Significant differences between sexes at the end of the study. p<0.05

-DBP (mm Hg): Low flavanone drink on women = B: 75.00 ± 1.60; F: 76.00 ± 1.60. NS High flavanone drink on women = B: 73.00 ± 1.70; F: 73.00 ± 1.60. NS Low flavanone drink on men = B: 81.00 ± 2.60; F: 82.00 ± 2.60. NS High flavanone drink on men = B: 79.00 ± 2.40; F: 79.00 ± 2.40. NS Significant differences between sexes at the end of the study. p<0.05

Constans J et al. Randomized, European (France) 25 subjects 600 mL/day of 0 mg/day in 600 4 weeks for one 5 days-food record Glucose, TC, Differences compared to baseline: (2015)S3 controlled, 51-56 years control drink (CD) mL/day of CD vs drink test + 5 weeks LDL-c, HDL-c, GLUCOSE METABOLISM: single blind, CVRF: vs 600 mL/day of 213 mg/day of for wash-out + 4 TG, Apo A-1, -Glucose (mmol/L): crossover, hypercholesterole orange juice (OJ) hesperidin in 600 weeks for the other Apo B, fibrinogen, CD = F: -0.27. NS chronic mic mL/day of OJ drink test + 5 weeks homocysteine, OJ = F: +0.06. NS for wash-out sVCAM-1, No significant differences between CD and OJ at the end of the study. sICAM-1, sE- selectin LIPID PROFILE: -TC (mmol/L): CD = F: -0.35. p<0.05 OJ = F: +0.20. NS No significant differences between CD and OJ at the end of the study.

-LDL-c (mmol/L): CD = F: -0.32. p<0.05 OJ = F: +0.11. NS No significant differences between CD and OJ at the end of the study.

-HDL-c (mmol/L): CD = F: -0.07. NS OJ = F: +0.00. NS No significant differences between CD and OJ at the end of the study.

-TG (mmol/L): CD = F: +0.10. NS OJ = F: +0.24. NS No significant differences between CD and OJ at the end of the study.

-Apo A-1 (mg/dL): CD = F: -3.00. NS OJ = F: +5.00. p<0.05 No significant differences between CD and OJ at the end of the study.

-Apo B (mg/dL): CD = F: -1.00. NS OJ = F: +8.00. p<0.05 No significant differences between CD and OJ at the end of the study.

COAGULATION BIOMARKERS: -Fibrinogen (mg/dL): CD = F: +11.00. NS OJ = F: +0.1. NS No significant differences between CD and OJ at the end of the study.

-Homocysteine (μmol/L): CD = F: +0.64. p<0.05 OJ = F: +0.33. NS No significant differences between CD and OJ at the end of the study.

INFLAMMATION BIOMARKERS: -sVCAM-1 (ng/mL): CD = F: 579.79 ± 50.23. NS OJ = F: 558.62 ± 47.36. NS No significant differences between CD and OJ at the end of the study.

-sICAM-1 (ng/mL): CD = F: 168.34 ± 11.03. NS OJ = F: 172.10 ± 11.70. NS No significant differences between CD and OJ at the end of the study.

-sE-selectin (ng/mL): CD= F: 121.90 ± 12.30. NS OJ = F: 110.12 ± 9.98. NS No significant differences between CD and OJ at the end of the study.

Rangel-Huerta Randomized, European (Spain) 100 subjects 500 mL/day of OJ vs 237 mg/day in 500 12 weeks for one BW, BMI, SBP, ANTHROPOMETRIC PARAMETERS: OD et al. double-blind, 18-65 years 500 mL/day of OJ mL/day of OJ vs intervention + 7 DBP, glucose, -BW (kg): (2015)S4 crossover, CVRF: with a high 582.50 mg/day of weeks for wash-out insulin, TC, LDL- OJ = B: 90.40 ± 1.50; F: 89.10 ± 1.50. p<0.05 chronic overweight or hesperidin hesperidin in 500 + 12 weeks for the c, HDL-c, TG, OJ with high hesperidin concentration = B: 90.60 ± 1.50; F: 88.80 ± 1.50. p<0.05 obese concentration mL/day of OJ other intervention Apo A-1, Apo B, No significant differences between OJ and OJ with high hesperidin concentration at oxLDL the end of the study.

-BMI (kg/m2): OJ = B: 32.50 ± 0.40; F: 32.00 ± 0.40. p<0.05 OJ with high hesperidin concentration = B: 32.60 ± 0.40; F: 31.90 ± 0.40. p<0.05 No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

VASCULAR PARAMETERS: -SBP (mm Hg): OJ = B: 128.00 ± 1.00; F: 124.00 ± 2.00. p<0.05 OJ with high hesperidin concentration = B: 127.00 ± 1.00; F: 124.00 ± 1.00. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-DBP (mm Hg): OJ = B: 79.00 ± 1.00; F: 76.00 ± 1.00. p<0.05 OJ with high hesperidin concentration = B: 78. 00 ± 1.00; F: 77.00 ± 1.00. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

GLUCOSE METABOLISM: -Glucose (mmol/L): OJ = B: 4.90 ± 0.10; F: 5.20 ± 0.10. p<0.05 OJ with high hesperidin concentration = 5.00 ± 1.00; F: 5.20 ± 0.01. p<0.05 Significant differences between OJ and OJ with high hesperidin concentration at the end of the study. p<0.05

-Insulin (µU/mL): OJ = B: 12.70 ±0.70; F: 11.50 ± 0.60. p<0.05 OJ with high hesperidin concentration = B: 13.80 ± 0.90; F: 12.70 ± 0.70. NS Significant differences between OJ and OJ with high hesperidin concentration at the end of the study. p<0.05

LIPID PROFILE: -TC (mmol/L): OJ = B: 5.60 ± 0.10; F: 5.60 ± 0.10. NS OJ with high hesperidin concentration = B: 5.60 ± 0.10; F: 5.60 ± 0.10. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-LDL-c (mmol/L): OJ = B: 3.39 ± 0.08; F: 3.47 ± 0.08. NS OJ with high hesperidin concentration = B: 3.41 ± 0.08. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-HDL-c (mmol/L): OJ = B: 1.29 ± 0.03; F: 1.32 ± 0.03. NS OJ with high hesperidin concentration = B: 1.32 ± 0.03; F: 1.29 ± 0.03. NSD No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-TG (mmol/L): OJ = B: 1.49 ± 0.07; F: 1.40 ± 0.07. p<0.05 OJ with high hesperidin concentration = B: 1.54 ± 0.07; F: 1.47 ± 0.06. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-Apo A-1 (mg/dL): OJ = B: 147.00 ± 2.00; F: 147.00 ± 2.00. NS OJ with high hesperidin concentration = B: 149.00 ± 2.00; F: 145.00 ± 2.00. p<0.05 No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

-Apo B (mg/dL): OJ = B: 95.00 ± 2.00; F: 91.00 ± 2.00. p<0.05 OJ with high hesperidin concentration = B: 96.00 ± 2.00; F: 93.00 ± 2.00. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

OXIDATION BIOMARKERS: -OxLDL (pg/mL): OJ = B: 335.00 ± 40.00; F: 343.00 ± 38.00. NS OJ with high hesperidin concentration = B: 322.00 ± 39.00; F: 326.00 ± 40.00. NS No significant differences between OJ and OJ with high hesperidin concentration at the end of the study.

Schär MY et al. Randomized, European (United 28 men 767 mL/day of CD 0 mg/day in CD vs 5 hours 24 hours-dietary SBP, DBP, RH- VASCULAR PARAMETERS: (2015)S5 placebo- Kingdom) 51-69 years (n=15) vs 767 mL of 320 mg/day of recall PAT index -SBP (mm Hg): controlled, CVRF: 10-20% of OJ (n=13) hesperidin in OJ CD = B: 128.20 ± 2.20; T5h: 123.60 ± 1.80. NS crossover, acute cardiovascular OJ = B: 126.30 ± 1.80; T5h: 123.60 ± 2.90. NS risk the next 10 No significant differences between control drink and OJ at the end of the study. years according to British -DBP (mm Hg): Hypertension CD = B: 80.20 ± 1.60; T5h: 75.10 ± 1.80. NS Society OJ = B: 77.90 ± 1.80; T5h: 73.60 ± 1.90. NS No significant differences between CD and OJ at the end of the study.

-RH-PAT index: CD = B: 2.78 ± 0.18; T5h: 2.66 ± 0.17. NS OJ= B: 2.77 ± 0.13; T5h: 2.68 ± 0.19. NS No significant differences between CD and OJ at the end of the study.

Buscemi S et al. Randomized, European (Italy) 31 subjects 500 mL/day of CD 0 mg/day in 1.5 weeks Food diary 24 hours FMD, GTN, IL-6, VASCULAR PARAMETERS: (2012)S6 placebo- 19-67 years (n=12) vs 500 placebo drink vs the day before the NOx -FMD (%): controlled, CVRF: metabolic mL/day of OJ (n=19) 159.50 mg/day of visit CD = B: 5.70 ± 2.40; F: 5.00 ± 1.80. NS single blind, syndrome hesperidin in 500 OJ = B: 5.70 ± 2.40; F: 7.90 ± 2.70. p<0.05 crossover, mL/day of OJ Significant differences between CD and OJ at the end of the study. p<0.05 chronic -GTN (%): CD = B: 17.90 ± 4.90; F: 17.90 ± 4.80. NS OJ = B: 17.90 ± 4.90; F: 18.80 ± 4.60. NS No significant differences between CD and OJ at the end of the study.

INFLAMMATION BIOMARKERS: -IL-6 (pg/mL): CD = B: 33.90 ± 2.60; F: 32.90 ± 3.20. NS OJ = B: 33.90 ± 2.60; F: 30.60 ± 2.60. p<0.05 Significant differences between CD and OJ at the end of the study. p<0.05

OXIDATION BIOMARKERS: -NOx (μmol/L): CD = B: 19.80 ± 4.60; F: 20.80 ± 6.20. NS OJ = B: 19.80 ± 4.60; F: 20.70 ± 7.20. NS No significant differences between CD and OJ at the end of the study.

Rizza S et al. Randomized, European (Italy) 24 subjects 1 placebo 0 mg/day in 3 weeks BMI, SBP, DBP, ANTHROPOMETRIC PARAMETERS: (2011)S7 placebo- 21-65 years capsule/day vs 1 placebo capsule vs FMD, GTN, TC, -BMI (kg/m2): controlled, CVRF: metabolic hesperidin 500 mg/day in 1 LDL-c, HDL-c, Placebo capsule = B: 34.70 ± 1.50; F: 34.70 ± 1.50. NS double-blind, syndrome capsule/day hesperidin capsule TG, Apo-A1, Hesperidin capsule = B: 34.70 ± 70.00; F: 34.70 ± 1.50. NS crossover, Apo-B, glucose, No significant differences between placebo capsule and hesperidin capsule at the end chronic insulin, QUICKI, of the study. fibrinogen, homocysteine, VASCULAR PARAMETERS: SAA protein, -SBP (mm Hg): VCAM, ICAM, Placebo capsule = B: 138.00 ± 3.00; F: 132.00 ± 2.00. NS sE-selectin Hesperidin capsule = B: 138.30 ± 3.00; F: 134.00 ± 3.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-DBP (mm Hg): Placebo capsule = B: 89.00 ± 2.00; F: 90.00 ± 2.00. NS Hesperidin capsule = B: 89.00 ± 2.00; F: 90.00 ± 2.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-FMD (%): Placebo capsule = B: 8.24 ± 0.88; F: 7.78 ± 0.76. NS Hesperidin capsule = B: 8.24 ± 0.88; F: 10.26 ± 1.19. p=0.05 Significant differences between placebo capsule and hesperidin capsule at the end of the study. p=0.02

-GTN (%) Placebo capsule = B: 13.98 ± 1.32; F: 14.40 ± 1.02. NS Hesperidin capsule = B: 13.98 ± 1.32; F: 14.04 ± 1.08. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

GLUCOSE METABOLISM: -Glucose (mmol/L): Placebo capsule = B: 7.33 ± 0.67; F: 7.16 ± 0.39. NS Hesperidin capsule = B: 7.33 ± 0.67; F: 6.99 ± 0.33. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-Insulin (μU/mL): Placebo capsule = B: 21.30 ± 2.10; F: 21.10 ± 1.90. NS Hesperidin capsule = B: 21.30 ± 2.10; F: 20.20 ± 2.10. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-QUICKI: Placebo capsule = B: 0.298 ± 0.004; F: 0.297 ± 0.003. NS Hesperidin capsule = B: 0.298 ± 0.004; F: 0.300 ± 0.004. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

LIPID PROFILE: -TC (mmol/L): Placebo capsule = B: 4.65 ± 0.21; F: 4.81 ± 0.21. NS Hesperidin capsule = B: 4.65 ± 0.21; F: 4.52 ± 0.21. NS Significant differences between placebo capsule and hesperidin capsule at the end of the study. p<0.05

-LDL-c (mmol/L): Placebo capsule = B: 3.09 ± 0.21; F: 3.17 ± 0.16. NS Hesperidin capsule = B: 3.09 ± 0.21; F: 2.99 ± 0.16. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study. -HDL-c (mmol/L): Placebo capsule = B: 0.96 ± 0.05; F: 0.88 ± 0.05. NS Hesperidin capsule = B: 0.96 ± 0.05; F: 0.91 ± 0.05. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-TG (mmol/L): Placebo capsule = B: 1.80 ± 0.13; F: 2.04 ± 0.19. NS Hesperidin capsule = B: 1.80 ± 0.13; F: 1.87 ± 0.11. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-Apo A1 (mg/dL): Placebo capsule = B: 134.00 ± 5.00; F: 136.00 ± 7.00. NS Hesperidin capsule = B: 134.00 ± 5.00; F: 137.00 ± 6.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-Apo B (mg/dL): Placebo capsule = B: 90.00 ± 4.00; F: 93.00 ± 4.00. NS Hesperidin capsule = B: 90.00 ± 4.00; F: 88.00 ± 4.00. NS Significant differences between placebo capsule and hesperidin capsule at the end of the study. p<0.05

COAGULATION BIOMARKERS: -Fibrinogen (mg/dL): Placebo capsule = B: 320.00 ± 14.00; F: 330.00 ± 16.00. NS Hesperidin capsule = B: 320.00 ± 14.00; F: 331.00 ± 15.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-Homocysteine (μM/L): Placebo capsule = B: 11.90 (10.30 – 14.90); F: 13.60 (10.60 – 16.70). NS Hesperidin capsule = B: 11.90 (10.30 – 14.90); F: 13.00 (10.20 – 15.50). NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

INFLAMMATION BIOMARKERS: -SAA protein (mg/L): Placebo capsule = B: 7.30 (5.60 – 6.10); F: 8.00 (5.60 – 11.20). NS Hesperidin capsule = B: 7.30 (5.60 – 6.10); F: 5.60 (3.20 – 7.80). NS Significant differences between placebo capsule and hesperidin capsule at the end of the study. p<0.05

-VCAM (ng/mL): Placebo capsule = B: 956.00 ± 29.00; F: 976.00 ± 30.00. NS Hesperidin capsule = B: 956.00 ± 29.00; F: 950.00 ± 27.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study. -ICAM (ng/mL): Placebo capsule = B: 291.00 ± 6.00; F: 299.00 ± 7.00. NS Hesperidin capsule = B: 291.60 ± 6.00; F: 294.00 ± 7.00. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-sE-selectin (ng/mL): Placebo capsule = B: 31.00 ± 2.00; F: 31.00 ± 2.00. NS Hesperidin capsule = B: 31.00 ± 2.00; F: 27.00 ± 2.00. NS Significant differences between placebo capsule and hesperidin capsule at the end of the study. p<0.05

Morand C et al. Randomized, European (France) 23 men 500 mL/day of CD 0 mg/day in 500 4 weeks 3 days-food record SBP, DBP, VASCULAR PARAMETERS: (2011)S8 placebo- 51-63 years and placebo mL/day of CD and glucose, insulin, -SBP (mm Hg): controlled, CVRF: supplement vs 500 1 placebo TC, LDL-c, HDL- CD and placebo supplement = F: 133.70 ± 2.10. NS crossover, overweight mL/day of CD and supplement vs 0 c, TG, IL-6, CD and placebo supplement vs CD and hesperidin supplement = F: +2.00 ± 2.90. NS chronic hesperidin mg/day in 500 sVCAM, sICAM- CD and placebo supplement vs OJ = F: -2.00 ± 2.80. NS supplement vs 500 mL/day of CD and 1, NOx No significant differences between the three interventions at the end of the study. mL/day of OJ 292 mg/day of pure hesperidin -DBP (mm Hg): capsule vs 292 CD and placebo supplement = F: 84.90 ± 2.10. NS mg/day of CD and placebo supplement vs CD and hesperidin supplement = F: -5.30 ± 2.00. hesperidin in 500 p<0.05 mL/day of OJ CD and placebo supplement vs OJ = F: -4.50 ± 2.00. p<0.05 Significant differences between the three interventions at the end of the study. p<0.023

GLUCOSE METABOLISM: -Glucose (mmol/L): CD and placebo supplement = F: 6.10 ± 0.20. NS CD and placebo supplement vs CD and hesperidin supplement = F: +0.20 ± 0.20. NS CD and placebo supplement vs OJ = F: +0.00 ± 0.20. NS No significant differences between the three interventions at the end of the study.

-Insulin (µU/mL): CD and placebo supplement = F: 16.90 ± 1.20. NS CD and placebo supplement vs CD and hesperidin supplement = F: -1.30 ± 1.50. NS CD and placebo supplement vs OJ = F: -1.20 ± 1.50. NS No significant differences between the three interventions at the end of the study.

LIPID PROFILE: -TC (mmol/L): CD and placebo supplement = F: 5.40 ± 0.20. NS CD and placebo supplement vs CD and hesperidin supplement = F: +0.40 ± 0.20. NS CD and placebo supplement vs OJ = F: +0.30 ± 0.20. NS No significant differences between the three interventions at the end of the study.

-LDL-c (mmol/L): CD and placebo supplement = F: 3.50 ± 0.20. NS CD and placebo supplement vs CD and hesperidin supplement = F: +0.40 ± 0.20. NS CD and placebo supplement vs OJ = F: +0.30 ± 0.20. NS No significant differences between the three interventions at the end of the study.

-HDL-c (mmol/L): CD and placebo supplement = F: 1.40 ± 0.10. NS CD and placebo supplement vs CD and hesperidin supplement = F: 0.00 ± 0.10. NS CD and placebo supplement vs OJ = F: 0.00 ± 0.10. NS No significant differences between the three interventions at the end of the study.

-TG (mmol/L): CD and placebo supplement = F: 1.30 ± 0.10. NS CD and placebo supplement vs CD and hesperidin supplement = F: +0.10 ± 0.10. NS CD and placebo supplement vs OJ = F: +0.10 ± 0.10. NS No significant differences between the three interventions at the end of the study.

INFLAMMATION BIOMARKERS: -IL-6 (pg/mL): CD and placebo supplement = F: 1.98 ± 0.25. NSD CD and placebo supplement vs CD and hesperidin supplement = F: -0.19 ± 0.29. NS CD and placebo supplement vs OJ = F: +0.11 ± 0.29. NS No significant differences between the three interventions at the end of the study.

-sVCAM-1 (ng/mL): CD and placebo supplement = F: 119.00 ± 119.00. NS CD and placebo supplement vs CD and hesperidin supplement = F: -283.00 ± 157.00. NSD CD and placebo supplement vs OJ = F: -302.00 ± 154.00. NS No significant differences between the three interventions at the end of the study.

-sICAM-1 (ng/mL): CD and placebo supplement = F: 360.00 ± 19.00. NS CD and placebo supplement vs CD and hesperidin supplement = F: +16.00 ± 27.00. NSD CD and placebo supplement vs OJ = F: +28.00 ± 26.00. NS No significant differences between the three interventions at the end of the study.

OXIDATION BIOMARKERS: -NOx (µmol/L): CD and placebo supplement = F: 37.90 ± 5.60. NS CD and placebo supplement vs CD and hesperidin supplement = F: -0.80 ± 7.00. NS CD and placebo supplement vs OJ = F: +13.50 ± 6.90. NS No significant differences between the three interventions at the end of the study.

Demonty I et al. Randomized, European 194 subjects 4 placebo 0 mg/day in 4 4 weeks Participants returned BW, BMI, TC, ANTHROPOMETRIC PARAMETERS: (2010)S9 placebo- (Holland) 51-69 years capsules/day vs 4 placebo capsules all used an unused LDL-c, HDL-c, -BW (kg): controlled, CVRF: hesperidin vs 800 mg/day in capsule boxes TG Placebo capsule = B: 74.00 ± 12.30; F: +0.09 ± 0.09. NS double blind, overweight, capsules/day 4 hesperidin Hesperidin capsule = B: 74.00 ± 9.50; F: +0.19 ± 0.10. NS parallel, chronic hypercholesterole capsules No significant differences between placebo capsule and hesperidin capsule at the end mic of the study.

-BMI (kg/m2): Placebo capsule = B: 25.10 ± 2.30; F: +0.02 ± 0.03. NS Hesperidin capsule = B: 25.10 ± 2.10; F: +0.07 ± 0.03. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

LIPID PROFILE: -TC (mmol/L): Placebo capsule = B: 6.18 ± 0.85; F: 6.22 ± 0.05. NS Hesperidin capsule = B: 6.18 ± 0.83; F: 6.19 ± 0.05. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-LDL-c (mmol/L): Placebo capsule = B: 3.97 ± 0.71; F: 4.00 ± 0.04. NSD Hesperidin capsule = B: 3.99 ± 0.77; F: 3.99 ± 0.04. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-HDL-c (mmol/L): Placebo capsule = B: 1.51 ± 0.45; F: 1.54 ± 0.02. NS Hesperidin capsule = B: 1.52 ± 0.40; F: 1.53 ± 0.02. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

-TG (mmol/L): Placebo capsule = B: 1.44 ± 0.60; F: 1.26 ± 0.03. NS Hesperidin capsule = B: 1.31 ± 0.54; F: 1.24 ± 0.04. NS No significant differences between placebo capsule and hesperidin capsule at the end of the study.

Aptekmann NP Randomized, South American 26 premenopausal 0 mL/day (n=13) vs 0 mg/day vs 54.60 13 weeks Self-report BW, BMI, body ANTHROPOMETRIC PARAMETERS: et al. (2010)S10 controlled, (Brazil) women 500 mL/day of OJ mg/day of fat, CT, LDL-c, -BW (kg): parallel, chronic 30-48 years (n=13) hesperetin in 500 HDL-c, TG Control group = B: 76.30 ± 15.30; F: 74.50 ± 15.90. p<0.05 CVRF: mL/day of OJ OJ group = B: 74.60 ± 13.00; F: ± 73.60 ± 12.40. p<0.05 overweight No significant differences between control group and OJ group at the end of the study.

-BMI (kg/m2): Control group = B: 29.00 ± 5.33; F: 28.30 ± 5.81. p<0.05 OJ group = B: 28.40 ± 4.46; F: 28.10 ± 4.47. p<0.05 No significant differences between control group and OJ group at the end of the study.

-Body fat (%): Control group = B: 39.30 ± 7.33; F: 33.80 ± 7.98. p<0.05 OJ group = B: 37.70 ± 7.56; F: 33.40 ± 7.42. p<0.05 No significant differences between control group and OJ group at the end of the study.

LIPID PROFILE: -TC (mmol/L): Control group = B: 5.03 ± 0.70; F: 4.95 ± 0.76. NS OJ group = B: 4.82 ± 0.74; F: 4.60 ± 0.74. p<0.05 No significant differences between control group and OJ group at the end of the study.

-LDL-c (mmol/L): Control group = B: 3.50 ± 0.87; F: 3.33 ± 0.85. NS OJ group = B: 3.03 ± 0.64; F: 2.59 ± 0.79. p<0.05 No significant differences between control group and OJ group at the end of the study.

-HDL-c (mmol/L): Control group = B: 1.53 ± 0.29; F: 1.44 ± 0.32. NS OJ group = B: 1.27 ± 0.28; F: 1.50 ± 0.31. p<0.05 No significant differences between control group and OJ group at the end of the study.

-TG (mmol/L): Control group = B: 1.02 ± 0.30; F: 0.95 ± 0.27. NS OJ group = B: 1.14 ± 0.49; F: 1.12 ± 0.32. NS No significant differences between control group and OJ group at the end of the study.

Cesar TB et al. Randomized, South American 22 subjects 0 mL/day (n=8) vs 0 mg/day vs 42 8.5 weeks BMI, TC, LDL-c, ANTHROPOMETRIC PARAMETERS: (2010)S11 controlled, (Brazil) 28-56 years 750 mL/day of OJ mg/day of HDL-c, TG -BMI (kg/m2): parallel, chronic CVRF: (n=14) hesperetin in 750 Control group = B: 30.00 ± 6.00; F: 30.00 ± 5.00. NS hypercholestero- mL/day of OJ OJ = B: 28.00 ± 5.00; F: 28.00 ± 5.00. NS lemic No significant differences between control group and OJ at the end of the study.

LIPID PROFILE: -TC (mmol/L): Control group = B: 5.51 ± 0.73; F: 5.75 ± 1.04. NS OJ = B: 5.95 ± 0.55; F: 5.49 ± 0.78. p<0.05 No significant differences between control group and OJ at the end of the study.

-LDL-c (mmol/L): Control group = B: 3.69 ± 0.68; F: 3.82 ± 0.94. NS OJ = B: 4.16 ± 0.44; F: 3.67 ± 0.68. p<0.05 No significant differences between control group and OJ at the end of the study.

-HDL-c (mmol/L): Control group = B: 1.14 ± 0.34; F: 1.14 ± 0.31. NS OJ = B: 1.12 ± 0.18; F: 1.17 ± 0.18. NS No significant differences between control group and OJ at the end of the study

-TG (mmol/L): Control group = B: 1.46 ± 0.60; F: 1.84 ± 0.73. p<0.05 OJ = B: 1.48 ± 0.49; F: 1.66 ± 0.59. NS No significant differences between control group and OJ at the end of the study.

CVRF, cardiovascular risk factors; SBP, systolic blood pressure; DBP, diastolic blood pressure; FMD, flow-mediated dilatation; QUICKI, quantitative insulin-sensitivity check index; TC, total cholesterol; LDL-c, low density lipoprotein cholesterol; HDL-c, high density lipoprotein cholesterol; TG, triglycerides; sVCAM-1, soluble vascular cell adhesion molecule-1; sICAM-1, soluble intercellular adhesion molecule-1; sE-selectin, soluble E-selectin; sP-selectin, soluble P-selectin; NS, no significant differences between basal and final values; CD, control drink; OJ, orange juice; Apo A- 1, apolipoprotein A-1; Apo B, apolipoprotein B; BW, body weight; BMI, body mass index; oxLDL, oxidized low density lipoprotein; RH-PAT index, reactive hyperemia-peripheral arterial tonometry index; NDA, no data available; GTN, glyceryl-nitrate dilation of the brachial artery; IL-6, interleukin 6; NOx, nitric oxide; SAA protein, serum amyloid A protein.

REFERENCES

S1. Salden BN, Troost FJ, de Groot E, et al. Randomized clinical trial on the efficacy of hesperidin 2S on validated cardiovascular biomarkers in healthy overweight individuals. Am J Clin Nutr.

2016;104(6):1523-1533.

S2. Kean RJ, Lamport DJ, Dodd GF, et al. Chronic consumption of flavanone-rich orange juice is associated with cognitive benefits : an 8-wk , randomized , double-blind , placebo-controlled trial in healthy older adults 1 – 3. Am J Clin Nutr. 2015;101:506-514.

S3. Constans J, Bennetau-Pelissero C, Martin JF, et al. Marked antioxidant effect of orange juice intake and its phytomicronutrients in a preliminary randomized cross-over trial on mild hypercholesterolemic men. Clin Nutr. 2015;34(6):1093-1100.

S4. Rangel-Huerta OD, Aguilera CM, Martin M V., et al. Normal or High Polyphenol Concentration in Orange Juice Affects Antioxidant Activity, Blood Pressure, and Body Weight in Obese or

Overweight Adults. J Nutr. 2015;145(8):1808-1816.

S5. Schar MY, Curtis PJ, Hazim S, et al. Orange juice-derived flavanone and phenolic metabolites do not acutely affect cardiovascular risk biomarkers: A randomized, placebo-controlled, crossover trial in men at moderate risk of cardiovascular disease1-5. Am J Clin Nutr. 2015;101(5):931-938.

S6. Buscemi S, Rosafio G, Arcoleo G, et al. Effects of red orange juice intake on endothelial function and inflammatory markers in adult subjects with increased. Am J Clin Nutr. 2012;95:1089-

1095.

S7. Rizza S, Muniyappa R, Iantorno M, et al. Citrus polyphenol hesperidin stimulates production of nitric oxide in endothelial cells while improving endothelial function and reducing inflammatory markers in patients with metabolic syndrome. J Clin Endocrinol Metab. 2011;96(5):782-792.

S8. Morand C, Dubray C, Milenkovic D, Lioger D, Franc J, Scalbert A. Hesperidin contributes to the vascular protective effects of orange juice : a randomized crossover study in healthy volunteers 1 – 3. Am J Clin Nutr. 2011;93(7):73-80.

S9. Demonty I, Lin Y, Zebregs YEMP, et al. The Citrus Flavonoids Hesperidin and Naringin Do Not Affect Serum Cholesterol in Moderately Hypercholesterolemic Men and Women. J Nutr.

2010;140(9):1615-1620.

S10. Aptekmann NP, Cesar TB. Orange juice improved lipid profile and blood lactate of overweight middle-aged women subjected to aerobic training. Maturitas. 2010;67(4):343-7.

S11. Cesar TB, Aptekmann NP, Araujo MP, Vinagre CC, Maranhão RC. Orange juice decreases low-density lipoprotein cholesterol in hypercholesterolemic subjects and improves lipid transfer to high-density lipoprotein in normal and hypercholesterolemic subjects. Nutr Res. 2010;30(10):689-694.

Projects 2, 3 and 4: Summarizing the CITRUS randomized controlled trial.

129

130

The CITRUS study

This section explains the RCT called CITRUS, a competitive project, from which 3 projects were realized and reported in the present thesis.

First, we will start with the evaluation of hesperidin consumption in OJ and hesperidin-enriched OJ (EOJ) in pre- and stage 1 hypertensive subjects on BP and PP (Project 2). Second, we evaluated the effects of hesperidin consumption in OJ and EOJ on the transcriptomic profile of PBMCs (Project 3). Finally, we evaluated the effects of hesperidin consumption in OJ and EOJ on plasma, serum, and urine metabolomic profiles (Project 4).

Study design: A randomized, parallel, double-blinded, and placebo-controlled clinical trial was performed. All the participants were randomly assigned to one of the three intervention groups, namely control drink (CD), OJ and EOJ, and they consume 500 mL/day of the corresponding drink for 12 weeks. Moreover, two single dose studies, one at the beginning of the study and the other one at the end of the study after 12 weeks, were performed. For single dose studies participants consumed 500 mL of the corresponding intervention drink in the postprandial state.

After 1 week with control dietary habits following nutritionist recommendations to limit the total intake of flavonoid-rich foods and citrus-containing foods and maintain their normal dietary habits, the participants started the clinical trial. Moreover, 48 hours before the two

131 single dose studies, the participants realized a phenolic compounds-free diet. During the sustained study the subjects attended 7 visits (V), and the two single dose studies were performed at V1 and V7. The first single dose study (V1) was realized to determine the postprandial effects of hesperidin in OJ, and the second single dose study (V7) was realized to assess the postprandial effects of hesperidin in OJ after the sustained consumption. At V1, V3, V5 and V7 a 3-day food record was obtained to determine their dietary habits, and blood and urine samples in fasting conditions were collected. Moreover, at each visit, a physical examination, physical activity questionnaire class and anthropometric measurements were performed.

Study population: From the subjects who attended to the preselection, 159 (53 women and 106 men) were included. Participants has pre- or stage 1 hypertension. Inclusion criteria were: - Age from 18 to 65 years old. - SBP ≥120 mm Hg. - No family history of CVDs or chronic diseases. - Willingness to provide informed consent before starting the study. Exclusion criteria were: - BMI ≥ 35 kg/m2. - Fasting glucose > 125 mg/dL. - SBP ≥ 160 mm Hg - DBP ≥ 100 mm Hg

132

- Taking antihypertensive medications. - Hyperlipemia or antilipemic medication. - Smoking. - Pregnancy or intending to become. - Use of medications, antioxidants, vitamin supplements or adherence to a vegetarian diet. - Chronic alcoholism. - Physical activity > 5 hours/week - Intestinal disorders - Anemia (hemoglobin ≤13 mg/dL in men and ≤12 mg/dL in women). - Consumption of a research product in the 30 day prior to inclusion in the present study. - Failure to follow the study guidelines.

The clinical trial was approved by the Clinical Research Ethical Committee of Hospital Sant Joan (14-12-18/12aclaassN1), Reus, Spain; was conducted in accordance to Helsinki Declaration and Good Clinical Practice Guidelines of the International Conference of Harmonization and were reported as CONSORT criteria. Finally, the clinical trial was registered at Clinical-Trials.gov: NCT02479568.

Intervention drinks: The three intervention drinks were supplied by the Florida Department of Citrus of the United States of America. The intervention drinks were CD with no hesperidin content, OJ containing 392 mg/500 mL hesperidin, and EOJ containing 670 mg/500 mL hesperidin. Ferrer HealthTech of Murcia (Spain) provided the micronized 2S hesperidin used to enrich the EOJ intervention because it is the form

133 that is naturally present in OJ and the most bioavailable.(103) The detailed composition of the three drinks is reported in Supporting Information Table S1 of the published version of Project 2.

Parameters measured: The following parameters were obtained and measured during the CITRUS study and reported in 3 different projects depending on the objective: -In the Project 2, the following were evaluated at V1, V3, V5 and V7:

o SBP and DBP levels, and pulse pressure (PP). o Serum levels of homocysteine, F2α-isoprostanes, ICAM-1, VCAM-1 and uric acid.

o For the single-dose study, SBP, DBP, PP and homocysteine were evaluated at baseline and after 2, 4 and 6 hours. -In the Project 3, the following were obtained to realize transcriptomic analysis at V1 and V7:

o Blood samples to obtain PBMCs. -In the Project 4, the following were obtained to realize the metabolomics analysis:

o Plasma samples at weeks 4, 8 and 12, and after 2, 4 and 6 hours of the single dose.

o Serum and urine samples at V1 and V7. o Serum samples were obtained at baseline and after 2, 4 and 6 hours of a single dose of hesperidin.

Finally, Figure 13 presents a schema of the CITRUS study that includes the 3 projects realized in subjects with pre- or stage 1 hypertension to evaluate the effects of hesperidin consumption in OJ and EOJ.

134

135

136

Project 2: Effects of hesperidin in orange juice on blood and pulse pressures in mildly hypertensive individuals: a randomized clinical trial (CITRUS study).

137

138

European Journal of Nutrition https://doi.org/10.1007/s00394-020-02279-0

ORIGINAL CONTRIBUTION

Efects of hesperidin in orange juice on blood and pulse pressures in mildly hypertensive individuals: a randomized controlled trial (Citrus study)

Rosa M. Valls1,2 · Anna Pedret1,2 · Lorena Calderón‑Pérez1,2 · Elisabet Llauradó1,2 · Laura Pla‑Pagà1,2 · Judit Companys1,2 · Ana Moragas3,4,5 · Francisco Martín‑Luján3,4,6 · Yolanda Ortega4,5,7 · Montse Giralt1 · Marta Romeu1 · Laura Rubió1,11 · Jordi Mayneris‑Perxachs2,8 · Núria Canela8 · Francesc Puiggrós2 · Antoni Caimari2 · Josep M. Del Bas2 · Lluís Arola2,9 · Rosa Solà1,2,10

Received: 18 December 2019 / Accepted: 11 May 2020 © The Author(s) 2020

Abstract Purpose To assess the sustained and acute efects, as well as the infuence of sustained consumption on the acute efects, of orange juice (OJ) with a natural hesperidin content and hesperidin-enriched OJ (EOJ) on blood (BP) and pulse (PP) pressures in pre- and stage-1 hypertensive individuals. Methods In a randomized, parallel, double-blind, placebo-controlled trial, participants (n = 159) received 500 mL/day of control drink, OJ, or EOJ for 12 weeks. Two dose–response studies were performed at baseline and after 12 weeks. Results A single EOJ dose (500 mL) reduced systolic BP (SBP) and PP, with greater changes after sustained treatment where a decrease in diastolic BP (DBP) also occurred (P < 0.05). SBP and PP decreased in a dose-dependent manner relative to the hesperidin content of the beverages throughout the 12 weeks (P < 0.05). OJ and EOJ decreased homocysteine levels at 12 weeks versus the control drink (P < 0.05). After 12 weeks of EOJ consumption, four genes related to hypertension (PTX3, NLRP3, NPSR1 and NAMPT) were diferentially expressed in peripheral blood mononuclear cells (P < 0.05). Conclusion Hesperidin in OJ reduces SBP and PP after sustained consumption, and after a single dose, the chronic consumption of EOJ enhances its postprandial efect. Decreases in systemic and transcriptomic biomarkers were concomitant with BP and PP changes. EOJ could be a useful co-adjuvant tool for BP and PP management in pre- and stage-1 hypertensive individuals.

Keywords Orange juice · Hesperidin · Blood pressure · Pulse pressure · Pre-hypertension

Introduction naringenin-7-O-rutinoside (narirutin) are the main citrus favanone components [2]. Flavonoid compounds are the most abundant phenolic Data from cohort studies reported an inverse associa- compounds in plants, and citrus favonoids, particularly tion between citrus fruit/favanone consumption and cer- present in orange juice (OJ), are attracting attention due to ebrovascular disease [3–5] and cardiovascular mortality their benefcial efects on cardiovascular risk factors [1]. [6–8]. Antihypertensive, antithrombotic, anti-infamma- OJ is a main dietary source of favanones, a subclass of tory, antilipemic, vasodilator, and antioxidant efects of favonoids, and hesperetin-7-O-rutinoside (hesperidin) and hesperidin have been reported in animal models [9–11]. Similar outcomes have been observed for narirutin [12]. Recently, hesperidin has been shown to reduce the ath- Electronic supplementary material The online version of this erosclerotic plaque area and macrophage foam cell forma- article (https://doi.org/10.1007/s00394-020-02279-0) contains tion in low density lipoprotein (LDL) receptor-defcient supplementary material, which is available to authorized users. mice [11]. The aforementioned properties of hesperidin * Anna Pedret have been considered to be the mechanisms responsible [email protected] for the benefcial efects of citrus favanone consumption Extended author information available on the last page of the article on cardiovascular disease in humans [13].

Vol.:(0123456789)1 3 European Journal of Nutrition

Concerning the antihypertensive efects of citrus fa- pregnant, use of medications, antioxidants, vitamin sup- vanones, data from animal studies showed that hespere- plements or adherence to a vegetarian diet, chronic alco- tin, a biological metabolite of hesperidin [10], exerts an holism, physical activity > 5 h/week, intestinal disorders, antihypertensive efect in hypertensive rats but not in anemia (hemoglobin ≤ 13 mg/dL in men and ≤ 12 mg/ normotensive rats [9, 14]. The antihypertensive efect, dL in women), consumption of a research product in the as well as vasodilatory and anti-infammatory activities, 30 days prior to inclusion in the study, or failure to follow has been reported to be mediated by the hesperetin-7-O- the study guidelines. Participants signed informed consent β-d-glucuronide conjugate [10]. In humans, a natural OJ, prior to their participation in the study, which was approved but not a hesperidin-enriched beverage, decreased blood by the Clinical Research Ethical Committee of HUSJ pressure (BP) in overweight and obese individuals [15]. (14-12-18/12aclaassN1), Reus, Spain. The protocol and trial Similarly, chronic consumption of hesperidin reduced BP were conducted in accordance with the Helsinki Declaration in type 2 diabetes patients [16], although no hypotensive and Good Clinical Practice Guidelines of the International efect was observed in healthy or overweight individuals Conference of Harmonization (GCP ICH) and were reported [17, 18]. In individuals at moderate risk of cardiovascular as CONSORT criteria. The trial was registered at Clinical- disease (CVD), no changes in BP or other cardiovascular Trials.gov: NCT02479568. risk biomarkers were observed after a single dose of OJ or a hesperidin supplement at 5 h post intake [19]. Among Intervention products favonoids subclasses, favone and favan-3-ol compounds, but not favanones, were related to the prevention of hyper- Intervention beverages (supplied by the Florida Department tension in a cohort of 87,242 women from the Nurses’ of Citrus, USA) were control drink (CD), an OJ containing Health Study [20]. 690 mg/L of hesperidin (the natural hesperidin content), Thus, data of the antihypertensive efect of hesperidin and an enriched orange juice (EOJ) containing 1200 mg/L consumption in humans remain controversial. Therefore, of hesperidin. Ferrer HealthTech (Murcia, Spain) provided we assessed both the sustained and acute efects, as well as the Micronized 2S Hesperidin used in EOJ enrichment. The the infuence of sustained consumption on acute efects, of 2S form, the one present naturally in the OJ, is the most real-life doses of OJ and a hesperidin-enriched dose on BP, bioavailable [18]. Beverages were analyzed for hesperidin pulse pressure (PP), and cardiovascular risk biomarkers in and narirutin content using chromatography–mass spectrom- pre- and stage-1 hypertensive individuals. Our hypothesis etry (LC–MS/MS) (Supporting Information Table S1). Daily was that hesperidin in OJ would provide benefts on BP doses of 500 mL of CD, OJ and EOJ, provided 0 mg/day, and PP not only after sustained consumption but also at 345 mg/day, and 600 mg/day of hesperidin, and 0 mg/day, postprandial level after a single dose. 64 mg/day, and 77.5 mg/day of narirutin, respectively. Inter- vention drinks were similar in appearance and smell, and were diferentiated only by a code assigned by an independ- Materials and methods ent researcher not related to the study to guarantee blinding. Flavanone contents of the OJ and the EOJ were stable Study population throughout the study.

Participants from the general population were recruited by means of news in the newspapers, social networks, and tab- Study design leaux advertisements in the Hospital Universitari Sant Joan (HUSJ)-Eurecat, Reus, Spain, between January 2016 and A randomized, parallel, double-blind, placebo-controlled June 2017. From 311 subjects assessed for eligibility, 159 clinical trial was performed (Supporting Information Fig. (53 women and 106 men) pre- or stage-1 hypertensive indi- S1). Participants were randomly assigned to one of the viduals, according to current guidelines [21], were recruited. three intervention groups—CD, OJ, or EOJ—to consume Inclusion criteria were as follows: age from 18 to 65, systolic 500 mL/day of the corresponding beverage for 12 weeks. blood pressure (SBP) ≥ 120 mmHg, no family history of car- Nested within the sustained consumption study were two diovascular disease or chronic disease, and willingness to dose–response studies, one at baseline and the other after provide informed consent before the initial screening visit. 12 weeks of sustained consumption, where the 500 mL/ Exclusion criteria were: body mass index (BMI) ≥ 35 kg/ dose was administered all at once and changes in the out- m2, fasting glucose > 125 mg/dL, SBP ≥ 160 mmHg and comes were recorded in the postprandial state. Participants diastolic blood pressure (DBP) > 100 mmHg or taking were randomly allocated to the three intervention groups antihypertensive medications, hyperlipemia or antilipemic by a computerized random-number generator made by an medication; smoking, pregnancy or intending to become independent statistician. PROC PLAN (SAS 9.2, Cary, NC:

1 3 European Journal of Nutrition

83 SAS Institute Inc.) with a 1:1:1 allocation using random Main outcome measures block sizes of 2, 4, and 6 was used. Participants, researchers and the statistician remained blinded to the type of product SBP and DBP were measured twice after 2–5 min of administered throughout the study. respite, with the patient in a seated position, with 1-min After enrolment and following a 1-week run-in period interval between, using an automatic sphygmomanometer with a control diet consisting of a maintained lifestyle and (OMRON HEM-907; Peroxfarma, Barcelona, Spain). The normal dietary habits based on nutritionist recommenda- mean values were used for statistical analyses. Ofce PP, tions, the participants started the intervention trial. How- which represents the force that the heart generates each time ever, during the intervention period, the participants were it contracts, was determined by the diference between SBP instructed to also maintain their dietary habits, to completely and DBP [25]. The main outcomes were measured in both refrain from consuming citrus-containing foods and to limit dose–response and sustained consumption studies. their total intake of favonoid-rich foods (tea, cofee, cocoa, wine and other fruit juices) to reduce the possible masking efects that can exert these foods on BP [22, 23]. During Secondary outcomes the sustained study, participants attended seven visits (V) at the HUSJ-Eurecat. Dose–response postprandial studies, per- Homocysteine in serum samples was determined by LC–MS/ formed at V1 and V7, lasted from 08:00 a.m. to 02:00 p.m., MS. F2α isoprostanes were determined by a quantitative and participants received a light meal before leaving. In sandwich enzyme-linked immunosorbent assay (ELISA) addition to the baseline (0 h), blood samples were collected (Caymanchem, MI, USA) in 24-h urine. Soluble Intercel- at 2 h, 4 h, and 6 h after the single dose of 500 mL. The lular Adhesion Molecule-1 (ICAM-1) and Soluble Vascu- adherence of the volunteers to their dietary habits through- lar Cell Adhesion Molecule-1 (VCAM-1) were determined out the study was assessed by a 3-day food record at V1, V3, in serum by the Luminex™xMAP technology with the V5, and V7. At each visit, subjects underwent a physical EPX010-40,232-901 kit eBioscience (Thermo Fisher Scien- examination by a general practitioner and completed a Physi- tifc, Waltham, Massachusetts, USA), in the the Bio-Plex™ cal Activity Questionnaire Class AF [24], and anthropomet- 200 instrument (Bio-Rad, Hercules, California, USA). Uric ric measurements were recorded. Samples were stored at acid was measured by standardized methods on an autoana- -80ºC in the central laboratory’s Biobanc of HUSJ-Eurecat lyzer 182 (Beckman Coulter-Synchron, Galway, Ireland) in ([email protected]) until required for batch analyses. serum samples. All biological biomarkers were measured in the sustained consumption study. Homocysteine was Compliance measures additionally measured after the single 500-mL dose of the corresponding intervention product in both dose–response The plasma levels of the following biomarkers of nutrient studies. exposures were measured by LC–MS/MS in the plasma samples: hesperetin-7-O-β-d-glucuronide, hesperetin-3-O- Transcriptomic analyses β-d-glucuronide, hesperetin-7-O-sulfate, naringin-4-O-β-d- glucuronide, naringin-glucuronide and naringin sulfate. The Gene expression was assessed in peripheral blood mono- extraction was carried out with a semi-automated process nuclear cells (PBMCs) with an Agilent Microarray Plat- using Agilent Bravo Automated Liquid Handling Platform. form (Agilent Technologies, Santa Clara, California, USA) Briefy, 20 μL of internal standard (Hesperetin d4) was in a subsample (n = 37) of participants (11, 15, and 11, mixed with 125 μL of plasma and 750 μL of methanol. The in CD, OJ, and EOJ groups, respectively) at baseline and mixture was vortexed and centrifuged at 4700 rpm at 4 °C, after 12 weeks. PBMC RNA was isolated using Ficoll and then 900 μL was evaporated in a Speed-Vac at room gradient separation GE Healthcare Bio Sciences, Barce- temperature. Residues were reconstituted in 25 μL of MeOH lona, Spain), RNA yield was quantifed with a Nanodrop and 75 μL of H 2O (1% of HFor) and injected in the LC–MS/ UV–VIS Spectrophotometer and integrity was measured MS, an Agilent 1200 series ultra-high-performance liquid with an Agilent 2100 Bioanalyzer using the Total RNA chromatography (UHPLC) system coupled to a 6490 Triple Nano kit and the Eukaryote Total RNA Nano (Agilent Quad mass spectrometer, with electrospray source ionization Technologies, Santa Clara, California, USA). Total RNA (ESI) operating in negative mode. from the PBMCs was labeled with one color (Cy3) (ref: 5190-2305, Agilent Technologies, Santa Clara, California, USA) and hybridized using a Gene Expression Hybridi- zation Kit (ref: 5188-5242, Agilent Technologies, Santa Clara, California, USA). Image scanning was performed with an Agilent Microarray Scanner System with SureScan

1 3 European Journal of Nutrition

High Resolution Technology (Agilent Technologies, Santa they presented a P value < 0.05 and a fold change ≤ −0.58 Clara, California, USA). Diferentially expressed genes or ≥ 0.58 in log2 scale (corresponding to 1.5-fold diference were subjected to functional and biochemical pathway in natural scale). Calculations were performed using the R analysis using Gene Ontology, Kyoto Encyclopedia of statistical language. Comparisons among treatments were Genes and Genomes (KEGG) (https://www.genome.jp/ carried out by an ANCOVA model adjusted by age and sex kegg) and PANTHER (protein annotation through evolu- and baseline values. Statistical signifcance was defned as a tionary relationship classifcation system (https://www. P value ≤ 0.05 for a two-sided test. Analyses were performed pantherdb.org/ ) [26] biochemical pathway databases. The using SPSS for Windows, version 21 (IBM corp., Armonk, analysis was performed using GeneCodis (https://www. NY, USA). All data were analyzed according to the pre- genecodis.dacya.ucm.es [27] software. specifed protocol. Selected genes related to hypertension were validated by PCR. Briefy, to analyze the expression of the genes and validate the DNA array results, cDNA was synthesized Results using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4 Barcelona, Spain) and MyGene Study participants Series Peltier Thermal Cycler (LongGene Scientific, Zhejiang, China) and used for reverse transcription. The Of the 311 subjects who were assessed for eligibility, 152 cDNA was subjected to quantitative reverse transcription- did not meet the inclusion criteria and were excluded. The polymerase chain reaction amplifcation using LightCycler remaining 159 participants were randomly allocated to 480 SYBR Green I Master (Roche Diagnostic, Sant Cugat the CD, OJ, and EOJ groups, (n = 53 in each group). Ulti- del Vallès, Barcelona, Spain) and a LightCycler 480 II mately, 129 participants completed the study (43 in the CD, system (Roche Diagnostic, Sant Cugat del Vallès, Barce- 46 in the OJ, and 40 in the EOJ groups) (Fig. 1). For the lona, Spain). dose–response study, of the 52 allocated participants, three discontinued the intervention from the beginning, and six were lost for the second dose–response study. Thus, 52 par- Sample size and power analyses ticipants (17 in the CD, 21 in the OJ, and 14 in the EOJ groups) were available for the frst dose–response study, and A sample size of 159 individuals was calculated assuming an 43 (13 in the CD, 18 in the OJ, and 12 in the EOJ groups) expected dropout rate of 20% and a type I error of 0.005 (two were available for the second dose–response study (Fig. 1). sided), which allows at least 80% power for the detection of No diferences in baseline characteristics were observed statistically signifcant diferences in the SBP of 4 mmHg among the groups (Supporting Information Table S2). The among the groups. The population standard deviation of the baseline characteristics of participants in the dose–response SBP was estimated to equal 6 mmHg [28]. study were similar to those of the whole sample. No changes in the level of physical activity were observed from the Statistical analyses beginning to the end of the study in any group (data not shown). No diferences in dietary intake were observed Descriptive data were expressed as the mean 95% confdence among groups with exception of protein (% energy) intake, interval (CI). The normality of variables was assessed by which was greater in the OJ group than in the EOJ one the Kolmogorov–Smirnov test. Non-parametric variables (P = 0.031) (Supporting Information Table S3). were log transformed. ANOVA was used to determine differences in baseline characteristics. Analyses were made by Compliance biomarkers intention-to-treat. Multiple imputation was made by linear regression analysis. Intra-treatment comparisons were per- The volunteer compliance intervention was considered formed by means of a general linear model with Bonferroni optimal because all the compliance biomarkers (hesperetin- correction and age and sex as covariables. Inter-treatment 7-O-β-d-glucuronide, hesperetin-3-O-β-d-glucuronide, comparisons were carried out by analysis of covariance hesperetin-7-O-sulfate, naringerine-4-O-β-d-glucuronide, (ANCOVA) model adjusted for age and sex. For transcrip- naringenin-glucuronide and naringenin-sulfate) increased tomic analyses, quality control was performed through signifcantly during the OJ and EOJ intervention compared principal component analyses. Statistical comparisons were with the baseline values and the CD group intervention. At performed by Student’s t test or Welch’s t test if proceeded. 12 weeks, the metabolite hesperetin-7-β-d-glucuronide was Multiple testing correction was performed using the Benja- the main diferentially expressed metabolite between the OJ mini–Hochberg False Discovery Rate (FDR) control proce- and EOJ groups and the CD group (P < 0.001). dure. Probes were assumed to be diferentially expressed if

1 3 European Journal of Nutrition

Fig. 1 Flow chart of the study

After 12 weeks of treatments (Fig. 2a), plasma hesperetin- was also observed (P = 0.056). After consumption of EOJ, 7-β-d-glucuronide increased in a dose-dependent manner SBP decreased in all evaluated weeks compared to CD, with the hesperidin content of the beverage administered the decreases reaching signifcance (P < 0.05) at all weeks, (P < 0.001 for linear trend), and the increase in the EOJ with the only exception of week 8 in which a borderline group was signifcantly higher than that of the OJ group signifcance (P = 0.078) was observed. The average of all (P < 0.05). In the dose–response studies, plasma hesperetin- decreases through the study was −6.35 and −7.36 mmHg 7-β-d-glucuronide increased at 4 and 6 h after OJ and EOJ for OJ and EOJ interventions, respectively. DBP decreased (P < 0.005 versus changes in CD), both at the beginning similarly after all interventions and in all weeks (P < 0.05) (Fig. 2b) and at the end of the study. The individual changes (data not shown). Changes in PP through the study are in plasma hesperetin-7-β-d-glucuronide are depicted in Sup- shown in Fig. 4. PP decreased in a dose-dependent manner plementary Fig. 4 . with the hesperidin content of the beverage administered (P < 0.05 for linear trend) in all weeks, but in the 12 week, Main outcomes the trend did not reach signifcance (P = 0.125). Concern- ing dose–response studies, at the beginning of the study Changes in SBP at 2, 6, 10 and 12 weeks are shown in Fig. 3. (Fig. 5a), signifcant decreases were observed in SBP at SBP decreased in a dose-dependent manner with the hes- 2 h and in PP at all evaluated times (P < 0.05) after a sin- peridin content of the beverage administered (P < 0.05 for gle dose of 500 mL only in the case of EOJ. No changes linear trend). SBP decreased at weeks 4, 8, and 12 after OJ were observed in DBP values. After 12 weeks of treatment consumption, the decreases reaching signifcance versus (Fig. 5b), a single dose of 500 mL resulted in changes in changes. BP and PP also only in the EOJ group (Panel B). DBP in the CD at week 4 and 12 (P < 0.05) by mean 95% decreased versus baseline at all evaluated times (P < 0.05), IC −5.58 (−9.8; −1.3) mmHg and −5.06 (−8.8; −1.3) and the decrease at 2 h and 6 h reached signifcance ver- mmHg, respectively. A borderline signifcance at week 8 sus changes in CD (P < 0.05). Additionally, the observed

1 3 European Journal of Nutrition

Fig. 2 Changes in plasma hesperitin-7-β-d-glucuronide after ingestion of control, orange juice (OJ), and enriched OJ. a After sustained consumption for 12 weeks (500 mL/day). b At the beginning of the study after a single dose of 500 mL. *P < 0.05 versus baseline; † ‡ ¥ P < 0.05 versus control group; P < 0.001 versus control; P < 0.05 Fig. 3 Changes in systolic blood pressure (SBP) at 2 (a), 6 (b), 10 (c), versus OJ and 12 (d) weeks after sustained consumption of control drink (CD), orange juice (OJ), and hesperidin-enriched OJ (EOJ). *P < 0.05 versus baseline; †P < 0.05 versus CD decreases in SBP at 6 h and those of PP at all evaluated times reached signifcance versus changes in CD (P < 0.05). consumption, uric acid concentrations were directly related At 4 h and 6 h postprandial after our, an inverse relationship to SBP, DBP, and PP (P < 0.05). No changes were observed was observed between hesperidin-7-β-d-glucuronide values in other secondary outcomes. After 12 weeks sustained con- and those of PP (R = −0.354, P = 0.023, and R = −0.377, sumption, the values of SBP directly correlated with those P = 0.015, respectively). At 6 h, an inverse relationship was of ICAM-1 (R = 0.251, P = 0.004) and VCAM (R = 0.185, also observed between the increase in hesperidin-7-β-d- P = 0.036) (Fig. S3 in the online-only Data Supplement), glucuronide and SBP values (R = −0.353, P = 0.024). and the decrease in F2-isoprostanes, although without signifcance, were directly correlated with the decreases in SBP Secondary outcomes (R = 0.178, P = 0.042). No adverse events were reported. All products were well At week 12 after sustained consumption, homocysteine tolerated. values decreased after OJ and EOJ treatments, and the In transcriptomic analyses, after the sustained consump- decreases reached signifcance versus changes after the CD tion study, four genes related to hypertension were identi- treatment. At this time point, homocysteine plasma values fed: Pentraxin-3 (PTX3); NLR family, pyrin domain con- decreased in the postprandial state at 2 h and 4 h after OJ and taining 3 (NLRP3); neuropeptide S receptor 1(NPSR1); and after 2 h of EOJ ingestion (P < 0.05) (Fig. S2 in the online- nicotinamide phosphoribosyl transferase (NAMPT), which only Data Supplement). Uric acid decreased at 12 weeks were diferentially expressed after 12 weeks of treatment. after EOJ treatment, reaching signifcance versus changes The expressions of the PTX3 and NAMPT genes decreased in the CD treatment (P = 0.044). ICAM-1 decreased at week signifcantly in PBMC after the EOJ intervention versus 12 after EOJ treatment (P = 0.032), but no changes were the control treatment (P < 0.05). Figure 6 shows the com- observed between treatments. At week 12 after sustained parisons among interventions considering the dot axis at

1 3 European Journal of Nutrition

also observed. Thus, these suggested that sustained consumption of hesperidin optimizes acute BP-lowering efects. After 12 weeks, sustained EOJ consumption-related decreases in uric acid and ICAM were observed. Homocyst- eine decreased at 12 weeks after OJ and EOJ, and postprandial decreases in homocysteine were also present after single doses of OJ and EOJ at the end of the study. In agreement with the decrease in SBP and PP, PTX3 and NAMPT gene expression decreased in PBMCs at 12 weeks after sustained EOJ treatment. Currently, the worldwide prevalence of hypertension exceeds 1.3 billion [29] and is the main risk factor for death and disability-adjusted life-years lost during 2010 [30]. A 10-mmHg SBP decrease is associated with reductions of 22 and 41% in coronary heart disease and stroke, respectively [31]. Decreases in SBP with medical therapies range from 5 to 15 mmHg [32]. The average reductions in SBP throughout our study were −6.35 and −7.36 mmHg for the OJ and EOJ interventions, respectively. Our data are in agreement with those obtained after 8–12 weeks of a treadmill exercise program in hypertensive individuals (6.2 mmHg) [33] and after consumption of the Dietary Approaches to Stop Hyper- tension (DASH) diet (6.74 mmHg) [34] as well as with the results of a meta-analysis reporting a mean reduction of 8 mmHg SBP by regular endurance exercise in hypertensive patients [35]. PP, a surrogate marker of aortic stifness, is recognized as a powerful and independent risk factor for Fig. 4 Changes in pulse pressure (PP) at 2 (a), 6 (b), 10 (c), and 12 (d) weeks after sustained consumption of control drink (CD), orange CVD with prognostic utility beyond BP measurements [36, juice (OJ), and hesperidin-enriched OJ (EOJ). *P < 0.05 versus base- 37]. Throughout our study, the average PP reductions was † ‡ line; P < 0.05 versus CD; P < 0.05 versus OJ −2.41 mmHg after EOJ. A 10-mmHg increase in PP is associated with a 13% increase in all-cause mortality and >20% increase in recurrent myocardial infarction [36]. P < 0.05 to be of signifcance. The decreases in SBP and Throughout the 12 weeks of sustained intervention with PP at week 12 were directly related to the decrease in PTX3 345 mg/day of hesperidin in OJ and 600 mg/day of hesperi- expression (R = 0.393, P = 0.016 and R = 0.487, P = 0.002, din EOJ, we observed decreases in SBP and PP, but not in respectively). The decrease in PP at week 12 was directly DBP. Our results are opposite of those reported in over- related to that of NAMPT expression (R = 0.344, P = 0.037). weight men, with a DBP-lowering efect but not an SBP Although no signifcance was observed, changes in NRLP3 one, after 4 weeks of OJ or a hesperidin-rich capsule pro- expression were inversely related to those of PP at week viding 292 mg and 146 mg of hesperidin/day, respectively 12 (R = −0.420, P = 0.010). In turn, expression of PTX3 at [23]. No benefts on BP were reported after sustained high week 12 was directly related to that of NAMPT (R = 0.759, hesperidin consumption of 549 mg/L/day over 8 weeks in P < 0.001). healthy elderly individuals [17]; 6 weeks at 420 mg/day in healthy volunteers [18]; or 3 weeks at 500 mg/day in individuals with metabolic syndrome [38]. In type 2 diabetes Discussion patients, however, consumption (500 mg/day) of hesperidin over 6 weeks led to decreases in SBP and DBP [16]. In the present study, on the one hand, SBP and PP decreased Diferences in populations and lengths of treatment could in a dose-dependent manner with the hesperidin content of account for discrepancies among studies. If the objective of the beverage administered throughout the 12 weeks of the a study is to improve a specifc cardiovascular risk factor, study. On the other hand, a single dose of 500-mL EOJ, but subjects that present symptoms associated with the specifc no other treatment, reduced SBP, and PP, greater changes cardiovascular risk factor should be included in the study when the dose was administered at the end of the study after [39]. Thus, the present study constitutes the frst RCT that 12 weeks of sustained treatment where DBP changes were assesses the efects of hesperidin on BP and PP in pre- and

1 3 European Journal of Nutrition

Fig. 5 Changes in systolic blood pressure (SBP), diastolic blood pressure (DBP), and pulse pressure (PP) after a single dose of 500 mL of control drink, orange juice (OJ), and enriched OJ at the beginning (a) and at the end of the study (b)a stage-1 hypertensive subjects; whereas, the populations stud- Accordingly, in our study changes in NLPR3 gene expres- ied in other RCTs were overweight, obese or diabetic with sion after 12 weeks were inversely associated with those no hypertension or elevated BP levels [40]. At present, few of PP. After 12 weeks of EOJ consumption, we observed reports exist concerning the efect of hesperidin on PP. Hes- a decrease in PBMC expression of two key hypertension- peridin reversed aortic stifness in mice [41]. Although the related genes: PTX3 and NAMPT. Serum levels of PTX3, a benefcial efect of favonoids on arterial stifness is emerg- marker of infammation activation, are elevated in hyperten- ing [42], our data are the frst available to supporting the sive patients [45], and experimental studies reported a direct efect of dietary favanones on human arterial stifness. role of PTX3 in vascular function and BP homeostasis [46]. Mechanisms by which hesperidin could contribute to NAMPT, also called visfatin, is secreted by visceral fat and the control of BP and PP are associated with improvements is a stimulator of proinfammatory cytokines [47]. NAMPT on endothelial function, oxidative stress, and infammation is elevated not only in hypertensive patients but also in pre- [8]. Homocysteine is associated with these risk factors and hypertensive patients [48, 49], leading to the proposal that with a renin–angiotensin system activation to induce a BP NAMPT is a marker for damage in the pre-hypertensive state increase [43]. In agreement with this, we observe a decrease [48]. Thus, in our study, the decrease in biochemical and in homocysteine concomitant with decreases in BP and PP transcriptomic markers could account for the decreases in after hesperidin treatments. After 12-week EOJ consump- BP and PP after intake of hesperidin-rich beverages. tion, ICAM-1 values decreased, and this decrease and those One factor that could minimize diferences among sus- of other infammatory and oxidative markers were directly tained hesperidin interventions could be the similar contri- related to the SBP decrease. In our work, plasma uric acid bution of the narirutin present in these treatments. In experi- decreased after 12-week EOJ consumption. Hyperuricemia mental and human studies, naringin and narirutin showed is strongly associated with hypertension and arterial stif- a hypotensive effect [12, 50, 51]. When comparing the ness through activation of the NLPR3 infammasome [44]. dose–response results on the main outcomes, however, only

1 3 European Journal of Nutrition

this study constitutes the frst human RCT that assessed a compliance marker, hesperitin-7-β-d-glucuronide metabolite, which is associated with PP and SBP values, and thus, these results add robustness to our study. Multiple measurements throughout the study permitted the assessment of the homogeneity of the results. One limitation is the inability to assess potential interactions between the interventions and other dietary components. Addition- ally, a larger sample size could have permitted detection of signifcant diferences between both hesperidin treatments. Although BP measurements were performed with maximal care, a 24-h ambulatory BP monitoring could have been more accurate. The fact that participants were pre- and stage-1 hypertensive individuals limits the extrapolation of the results to the general population. Whether additional or diferent efects would have been observed over longer time periods is unknown, but longer intervention periods could have afected the compliance of the individuals. In summary, our results show that the intake of hesperidin in OJ decreases SBP and PP after sustained consumption in Fig. 6 Comparisons of changes in gene expression (log2FC, mean (95%CI) among treatments after interventions. PTX3, Pentraxin-3; a dose-dependent manner with the hesperidin content of the NLRP3, NLR family, pyrin domain containing 3; NPSR1, neuropep- beverage administered. Chronic consumption of hesperidin- tide S receptor 1; and NAMPT, nicotinamide phosphoribosyl trans- rich OJ enhances the postprandial response of decreasing ferase. The dot axis displays the signifcance between orange juices at SBP, DBP and PP. Decreases in homocysteine, uric acid the P < 0.05 level. *P < 0.05 and infammatory markers at the systemic level and in PTX3 and NAMPT at the transcriptomic level could account for the the EOJ single-dose intervention was capable of decreasing observed changes in BP and PP. SBP, DBP, and PP at the postprandial level. At present, few reports exist concerning the dose–response efect of hes- Perspectives peridin consumption in humans. No changes in SBP or DBP have been reported at 5 h after OJ or hesperidin supplemen- In a randomized, controlled clinical trial with pre- and tation (containing 320 mg of hesperidin) in healthy elderly stage-1 hypertensive individuals, we showed that sustained individuals, despite an increase in hesperidin at this time consumption of hesperidin promoted a dose-dependent point [19]. The fact that a unique measurement was obtained decrease in SBP and PP with the hesperidin content of the after a single dose could explain diferences between studies. beverage administered. Our data are the frst to support an To the best of our knowledge, our data are the frst to report efect of dietary favanones on human arterial stifness. the postprandial benefts of a hesperidin-enriched beverage Additionally, we report for the frst time the postprandial to BP and PP, as well as the fact that its sustained consump- benefts of a hesperidin-enriched beverage to BP and PP, tion enhances these benefts. as well as the fact that its sustained consumption enhances The study has strengths and limitations. As a strength, these benefts. Regular consumption of OJ, particularly the participants’ diets were monitored throughout the entire hesperidin-rich OJ, could be a useful co-adjuvant tool for study, and avoiding hesperidin intake and limiting the con- BP management in pre- and stage-1 hypertensive individu- sumption of flavonoid-rich foods were given as dietary als. This fact has public health implications in preventive recommendations to all the participants, which is of spe- medicine for reducing the secondary efects of long-term cial interest in nutritional RCTs because these guidelines medical treatment of mild hypertension. would limit confounding between other dietary compounds and the dietary intervention [40]. The dietary recommen- Acknowledgements The authors thank Ferrer HealthTech (Murcia, dations were established equally for all the intervention Spain) for donating the “Micronized 2S Hesperidin” to the Department of Citrus for the study. The Florida Department of Citrus, an execu- groups (CD, OJ and EOJ), and thus, the possible changes in tive agency of the state of Florida, USA, provided a grant and study the metabolome profle and consequently the downstream drinks to the Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutri- efects on BP due to these dietary modifcations would be ció i Salut, Reus, Spain. We thank the following for their enthusiastic equally observed in all the groups, which would result in the support in the conduct of the study: Miguel Querol, Lorena Torrado, Roger Esteve, Anabel Quesada, Davinia Ribes, Joana Contreras, Sandra control of these changes. Another important strength is that

1 3 European Journal of Nutrition

Bages, and Maria-Isabel Covas (NUPROAS HB). The NFOC-Salut 6. Mink PJ, Scraford CG, Barraj LM et al (2007) Flavonoid intake group is a consolidated research group of Generalitat de Catalunya, and cardiovascular disease mortality: a prospective study in post- Spain (2017 SGR522). menopausal women. Am J Clin Nutr 85:895–909. https://doi. org/10.1093/ajcn/85.3.895 Author contributions RMV, FP, AC, JPdB, LLA, RS: designed the 7. Wang X, Ouyang YY, Liu J, Zhao G (2014) Flavonoid intake and research (project conception, development of overall research plan, risk of CVD: a systematic review and meta-analysis of prospective and study oversight); RMV, AP, LC, ELL, LPP, JC, AM, FML, YO, cohort studies. Br J Nutr 111:1–11. https://doi.org/10.1017/S0007 MG, MR, LR, JMP, NC, AC, RS: conducted the research (hands-on 11451300278X performance of the experiments and data collection); RMV, AP, LC, 8. Cassidy A, Bertoia M, Chiuve S et al (2016) Habitual intake of ELL, LPP, JMP, NC: provided essential reagents or provided essential anthocyanins and favanones and risk of cardiovascular disease materials (applies to authors who contributed by providing animals, in men. Am J Clin Nutr 104:587–594. https://doi.org/10.3945/ constructs, databases, etc., necessary for the research); RMV, AP, MR, ajcn.116.133132 JMP, NC, FP, AC, JMdB, RS: analysed data or performed statistical 9. Ohtsuki K, Abe A, Mitsuzumi H et al (2002) Efects of long-term analysis; RMV, AP, LC, ELL, LPP, JC, JMP, NC, FP, AC, JMdB, LLA, administration of hesperidin and glucosyl hesperidin to spontane- RS: wrote the paper (only authors who made a major contribution); ously hypertensive rats. J NutrSci Vitaminol 48:420–422. https:// RMV, AP, LLA, RS: had primary responsibility for fnal content. doi.org/10.3177/jnsv.48.420 10. Yamamoto M, Jokura H, Hashizume K et al (2013) Hesperidin metabolite hesperetin-7-O-glucuronide, but not hesperetin-3′-O- Funding The Florida Department of Citrus, an executive agency of glucuronide, exerts hypotensive, vasodilatory, and anti-infamma- the state of Florida, USA, provided a grant and study drinks to the tory activities. Food Funct 4:1346–1351. https://doi.org/10.1039/ Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, c3fo60030k Reus, Spain. A.P. has a Torres Quevedo contract (Subprograma Estatal 11. Sun YZ, Chen JF, Shen LM et al (2017) Anti-atherosclerotic de Incorporación, Plan Estatal de Investigación Científca y Técnica efect of hesperidin in LDLr−/−mice and its possible mecha- y de Innovación). The funding bodies had no role in study design, nism. Eur J Pharmacol 815:109–117. https://doi.org/10.1016/j. data collection and analysis, decision to publish, or preparation of the ejphar.2017.09.010 manuscript. 12. Ikemura M, Sasaki Y, Giddings JC, Yamamoto J (2012) Preven- tive efects of hesperidin, glucosyl hesperidin and naringin on Compliance with ethical standards hypertension and cerebral thrombosis in stroke-prone spontaneously hypertensive Rats. Phyther Res 26:1272–1277. https://doi. Conflict of interest The authors declare no confict of interest. org/10.1002/ptr.3724 13. Testai L, Calderone V (2017) Nutraceutical value of citrus favanones and their implications in cardiovascular disease. Nutri- Open Access This article is licensed under a Creative Commons Attri- ents 9:E502. https://doi.org/10.3390/nu9050502 bution 4.0 International License, which permits use, sharing, adapta- 14. Guirro M, Costa A, Gual-Grau A et al (2018) Multi-omics tion, distribution and reproduction in any medium or format, as long approach to elucidate the gut microbiota activity: Metaproteom- as you give appropriate credit to the original author(s) and the source, ics and metagenomics connection. Electrophoresis 39:1692–1701. provide a link to the Creative Commons licence, and indicate if changes https://doi.org/10.1002/elps.201700476 were made. The images or other third party material in this article are 15. Rangel-Huerta OD, Aguilera CM, Martin MV et al (2015) Normal included in the article’s Creative Commons licence, unless indicated or high polyphenol concentration in orange juice afects antioxi- otherwise in a credit line to the material. If material is not included in dant activity, blood pressure, and body weight in obese or over- the article’s Creative Commons licence and your intended use is not weight adults. J Nutr 145:1808–1816. https://doi.org/10.3945/ permitted by statutory regulation or exceeds the permitted use, you will jn.115.213660 need to obtain permission directly from the copyright holder. To view a 16. Homayouni F, Haidari F, Hedayati M et al (2018) Blood pres- copy of this licence, visit http://creativeco mmons .org/licen ses/by/4.0/ . sure lowering and anti-infammatory efects of hesperidin in type 2 diabetes; a randomized double-blind controlled clinical trial. Phyther Res 32:1073–1079. https://doi.org/10.1002/ptr.6046 References 17. Kean RJ, Lamport DJ, Dodd GF et al (2015) Chronic consumption of favanone-rich orange juice is associated with cognitive benefts: An 8-wk, randomized, double-blind, placebo-controlled 1. Li C, Schluesener H (2017) Health-promoting efects of the citrus trial in healthy older adults. Am J Clin Nutr 101:506–514. https:// favanone hesperidin. Crit Rev Food Sci Nutr 57:613–631. https doi.org/10.3945/ajcn.114.088518 ://doi.org/10.1080/10408398.2014.906382 18. Salden BN, Troost FJ, De Groot E et al (2016) Randomized clini- 2. Crozier A, Cliford MN, Ashihara H (2006) Plant secondary cal trial on the efcacy of hesperidin 2S on validated cardiovas- metabolites: occurrence, structure and role in the human diet. cular biomarkers in healthy overweight individuals 1,2. Am J Clin Blackwell Publisher, Hoboken Nutr 104:1523–1533. https://doi.org/10.3945/ajcn.116.136960 3. Knekt P, Kumpulainen J, Järvinen R et al (2002) Flavonoid intake 19. Schar MY, Curtis PJ, Hazim S et al (2015) Orange juice-derived and risk of chronic diseases. Am J Clin Nutr 76:560–568. https:// favanone and phenolic metabolites do not acutely afect car- doi.org/10.1093/ajcn/76.3.560 diovascular risk biomarkers: a randomized, placebo-controlled, 4. Cassidy A, Rimm EB, O’Reilly ÉJ et al (2012) Dietary favo- crossover trial in men at moderate risk of cardiovascular dis- noids and risk of stroke in women. Stroke 43:946–951. https:// ease1-5. Am J Clin Nutr 101:931–938. https://doi.org/10.3945/ doi.org/10.1161/STROKEAHA.111.637835 ajcn.114.104364 5. Johnsen SP, Overvad K, Stripp C et al (2003) Intake of fruit and 20. Cassidy A, O’Reilly ÉJ, Kay C et al (2011) Habitual intake of vegetables and the risk of ischemic stroke in a cohort of Dan- favonoid subclasses and incident hypertension in adults. Am J ish men and women. Am J Clin Nutr 78:57–64. https://doi. Clin Nutr 93:338–347. https://doi.org/10.3945/ajcn.110.006783 org/10.1093/ajcn/78.1.57 21. Piepoli MF, Hoes AW, Agewall S et al (2016) 2016 European Guidelines on cardiovascular disease prevention in clinical

1 3 European Journal of Nutrition

practice. Eur Heart J 37:2315–2381. https://doi.org/10.1093/eurhe the REACH Registry. J Am Coll Cardiol 67:392–403. https://doi. artj/ehw106 org/10.1016/j.jacc.2015.10.084 22. Hooper L, Kroon PA, Rimm EB et al (2008) Flavonoids, favo- 38. Rizza S, Muniyappa R, Iantorno M et al (2011) Citrus polyphe- noid-rich foods, and cardiovascular risk: a meta-analysis of ran- nol hesperidin stimulates production of nitric oxide in endothe- domized controlled trials. Am J Clin Nutr 88:38–50. https://doi. lial cells while improving endothelial function and reducing org/10.1093/ajcn/88.1.38 infammatory markers in patients with metabolic syndrome. J 23. Morand C, Dubray C, Milenkovic D et al (2011) Hesperidin Clin Endocrinol Metab 96:E782–E792. https://doi.org/10.1210/ contributes to the vascular protective efects of orange juice: a jc.2010-2879 randomized crossover study in healthy volunteers. Am J Nutr 39. Brown L, Rosner B, Willett WW, Sacks FM (1999) Cholesterol- 93:73–80. https://doi.org/10.3945/ajcn.110.004945 lowering efects of dietary fber: a meta-analysis. Am J Clin Nutr 24. Vallbona Calbó C, Roure Cuspinera E, Violan Fors M (2007) Guia 69:30–42. https://doi.org/10.1093/ajcn/70.5.942 de Prescripció de l’Exercici Físic per a la Salut (Guia PEFS) 40. Pla-Pagà L, Companys J, Calderón-Pérez L et al (2019) Efects 25. Siu AL, U.S. Preventive Services Task Force (2015) Screening for of hesperidin consumption on cardiovascular risk biomarkers: a high blood pressure in adults: U.S. Preventive Services Task Force systematic review of animal studies and human randomized clini- Recommendation Statement. Ann Intern Med 163:778. https://doi. cal trials. Nutr Rev 77:845–864. https://doi.org/10.1093/nutrit/ org/10.7326/M15-2223 nuz036 26. Mi H, Huang X, Muruganujan A et al (2017) PANTHER version 41. Ouyang A, Garner TB, Fleenor BS (2017) Hesperidin reverses 11: Expanded annotation data from Gene Ontology and Reactome perivascular adipose-mediated aortic stifness with aging. Exp pathways, and data analysis tool enhancements. Nucleic Acids Res Gerontol 97:68–72. https://doi.org/10.1016/j.exger.2017.08.003 45:D183–D189. https://doi.org/10.1093/nar/gkw1138 42. Lilamand M, Kelaiditi E, Guyonnet S et al (2014) Flavonoids and 27. Nogales-Cadenas R, Carmona-Saez P, Vazquez M et al (2009) arterial stifness: promising perspectives. Nutr Metab Cardiovasc GeneCodis: Interpreting gene lists through enrichment analysis Dis 24:698–704. https://doi.org/10.1016/j.numecd.2014.01.015 and integration of diverse biological information. Nucleic Acids 43. Wu H, Wang B, Ban Q et al (2018) Association of total homo- Res 37:W317–W322. https://doi.org/10.1093/nar/gkp416 cysteine with blood pressure in a general population of Chinese 28. Lawes CMM, Vander HS, Law MR et al (2000) High blood pres- adults: a cross-sectional study in Jiangsu province. China BMJ sure. World Health Organization, Geneva Open 8:e021103. https://doi.org/10.1136/bmjopen-2017-021103 29. Bloch MJ (2016) Worldwide prevalence of hypertension exceeds 44. Kuwabara M, Kanbay M, Hisatome I (2018) Uric acid and hyper- 1.3 billion. J Am Soc Hypertens 10:753–754. https://doi. tension because of arterial stifness. Hypertension 72:582–584. org/10.1016/j.jash.2016.08.006 https://doi.org/10.1161/HYPERTENSIONAHA.118.11496 30. Lim SS, Vos T, Flaxman AD et al (2012) A comparative risk 45. Parlak A, Aydogan U, Iyisoy A et al (2012) Elevated pentraxin-3 assessment of burden of disease and injury attributable to 67 levels are related to blood pressure levels in hypertensive patients: risk factors and risk factor clusters in 21 regions, 1990–2010: an observational study. Anadolu Kardiyol Dergisi/Anatol J Car- A systematic analysis for the Global Burden of Disease Study diol 12:298–304. https://doi.org/10.5152/akd.2012.092 2010. Lancet 380:2224–2260. https://doi.org/10.1016/S0140 46. Carrizzo A, Lenzi P, Procaccini C et al (2015) Pentraxin 3 induces -6736(12)61766-8 vascular endothelial dysfunction through a P-selectin/matrix met- 31. Law MR, Morris JK, Wald NJ (2009) Use of blood pressure low- alloproteinase-1 pathway. Circulation 131:1495–1505. https://doi. ering drugs in the prevention of cardiovascular disease: Meta- org/10.1161/CIRCULATIONAHA.114.014822 analysis of 147 randomised trials in the context of expectations 47. Moschen AR, Kaser A, Enrich B et al (2007) Visfatin, an adipo- from prospective epidemiological studies. BMJ. https://doi. cytokine with proinfammatory and immunomodulating proper- org/10.1136/bmj.b1665 ties. J Immunol 178:1748–1758. https://doi.org/10.4049/jimmu 32. Morgan TO, Anderson AIE, MacInnis RJ (2001) ACE inhibitors, nol.178.3.1748 beta-blockers, calcium blockers, and diuretics for the control of 48. Gunes F, Akbal E, Cakir E et al (2012) Visfatin may be a novel systolic hypertension. Am J Hypertens 14:241–247. https://doi. marker for identifying stages of essential hypertension in advanced org/10.1016/S0895-7061(00)01266-8 age patients. Intern Med 51:553–557. https://doi.org/10.2169/inter 33. Pagonas N, Dimeo F, Bauer F et al (2014) The impact of aerobic nalmedicine.51.6609 exercise on blood pressure variability. J Hum Hypertens 28:367– 49. Liakos CI, Sanidas EA, Perrea DN et al (2016) Apelin and Vis- 371. https://doi.org/10.1038/jhh.2013.121 fatin plasma levels in healthy individuals with high normal blood 34. Saneei P, Salehi-Abargouei A, Esmaillzadeh A, Azadbakht L pressure. Am J Hypertens 29:549–552. https://doi.org/10.1093/ (2014) Infuence of Dietary Approaches to Stop Hypertension ajh/hpv136 (DASH) diet on blood pressure: a systematic review and meta- 50. Habauzit V, Verny MA, Milenkovic D et al (2015) Flavanones analysis on randomized controlled trials. Nutr Metab Cardiovasc protect from arterial stifness in postmenopausal women consum- Dis 24:1253–1261. https://doi.org/10.1016/j.numec d.2014.06.008 ing grapefruit juice for 6 mo: a randomized, controlled, crosso- 35. Cornelissen VA, Buys R, Smart NA (2013) Endurance exercise ver trial. Am J Clin Nutr 102:66–74. https://doi.org/10.3945/ beneficially affects ambulatory blood pressure: a systematic ajcn.114.104646 review and meta-analysis. J Hypertens 31:639–648. https://doi. 51. Reshef N, Hayari Y, Goren C et al (2005) Antihypertensive org/10.1097/HJH.0b013e32835ca964 efect of sweetie fruit in patients with stage I hypertension. Am 36. Harbaoui B, Nanchen D, Lantelme P et al (2018) Prognostic value J Hypertens 18:1360–1363. https://doi.org/10.1016/j.amjhy of pulse pressure after an acute coronary syndrome. Atheroscle- per.2005.05.021 rosis. https://doi.org/10.1016/j.atherosclerosis.2018.07.013 37. Selvaraj S, Steg PG, Elbez Y et al (2016) Pulse pressure and risk for cardiovascular events in patients with atherothrombosis from

1 3 European Journal of Nutrition

Afliations

Laura Rubió 7 Primary Care Centre Salou, Institut Català de la Salut, [email protected] Tarragona, Spain 8 Eurecat, Centre Tecnològic de Catalunya, Centre for Omic 1 Functional Nutrition, Oxidation, and Cardiovascular Sciences, Reus, Spain Diseases Group (NFOC‑Salut), Universitat Rovira i Virgili, Facultat de Medicina i Ciències de La Salut, Reus, Spain 9 Departament de Bioquímica i Biotecnologia, Grup de Recerca en Nutrigenòmica, Universitat Rovira i Virgili, 2 Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició Tarragona, Spain i Salut, Av. de La Universitat, 1, 43204 Reus, Spain 10 Hospital Universitari Sant Joan de Reus, Reus, Spain 3 Departament de Medicina i Cirurgia, Universitat Rovira i Virgili, Reus, Spain 11 Present Address: Food Technology Department, XaRTA-TPV, Agrotecnio Center, Escola Tècnica Superior 4 Institut Universitari d’Investigació en Atenció d’Enginyeria Agrària, University of Lleida, Av/ Alcalde Primària-IDIAP Jordi Gol, Tarragona, Spain Rovira Roure 191, 25198 Lleida, Spain 5 Primary Care Centre Jaume I, Institut Català de la Salut, Tarragona, Spain 6 Primary Care Centre El Morell, Institut Català de la Salut, Tarragona, Spain

1 3 Online Supplemental Tables for the manuscript

Effects of hesperidin in orange juice on blood and pulse pressures in mildly hypertensive individuals: a randomized controlled trial. Rosa M. Valls1,2, Anna Pedret2,1*, Lorena Calderón-Pérez1,2, Elisabet Llauradó1,2, Laura Pla-Pagà2,1, Judit Companys2,1, Ana Moragas3,4,5, Francisco Martín-Luján3,4,6, Yolanda Ortega4,5,7, Montse Giralt1, Marta Romeu1, Laura Rubió1, Jordi Mayneris-Perxachs2,8, Núria Canela8, Francesc Puiggròs2, Antoni Caimari2, Josep M. Del Bas2, Lluís Arola2,9, Rosa Solà1,2,10.

1Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut, Functional Nutrition, Oxidation, and Cardiovascular Diseases Group (NFOC- Salut), Reus, Spain. 2Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain. 3Universitat Rovira i Virgili, Departament de Medicina i Cirurgia, Reus, Spain 4Institut Universitari d’Investigació en Atenció Primària-IDIAP Jordi Gol, Tarragona, Spain. 5Primary Care Centre Jaume I, Institut Català de la Salut, Tarragona, Spain. 6Primary Care Centre El Morell, Institut Català de la Salut, Tarragona, Spain. 7Primary Care Centre Salou, Institut Català de la Salut, Tarragona, Spain. 8Eurecat, Centre Tecnològic de Catalunya, Centre for Omic Sciences, Reus, Spain 9Universitat Rovira i Virgili, Departament de Bioquímica i Biotecnologia, Grup de Recerca en Nutrigenòmica, Tarragona, Spain. 10Hospital Universitari Sant Joan de Reus, Reus, Spain.

*Corresponding Author: Anna Pedret, PhD. Eurecat, Centre Tecnològic de Catalunya. Unitat de Nutrició i Salut. Av. de la Universitat, 1, 43204 Reus, Spain. Tel: (+34) 977 75 14 84. E-mail: [email protected]

151

Table S1. Composition of the intervention products calculated by 500 mL/day* Intervention

CD OJ EOJ

Acidity, % 2.49 2.49 2.49

Sugar, g 43.1 37.7 37.7

Vitamin C, mg 235.3 235.3 235.3

Citric acid, g 3.40 3.40 3.40

Narirutin, mg ND 64 77

Hesperidin, mg ND 392 670

*In frozen concentrated canned drink, once diluted 3.4:1 (water to syrup). Abbreviations: CD, control drink; OJ, orange juice; EOJ, enriched orange juice; kcal, kilocalories; ND, non-detectable.

152

Table S2. Baseline characteristics of participants by intervention group Variable CD OJ EOJ P (n = 53) (n= 53) (n= 53) Age, y 45.4 ±13.0 43.3 ± 43.6 ± 0.629 12.0 11.8 Females, % 34.0 32.1 34.0 0.981 SBP, mm Hg 132 ± 9.94 132 ± 134 ± 9.82 0.687 9.11 DPB, mm Hg 79 ± 8.14 80 ± 8.42 79 ± 10.2 0.868 Pulse pressure, mm 53 ± 9.09 52 ± 8.05 54 ± 6.74 0.261 Hg Weight, kg 77.3 ± 78.8 ± 75.9 ± 11.6 0.523 15.4 12.2 BMI, kg/m2 26.1 ± 3.8 26.4 ± 3.6 26.1 ± 3.3 0.858 Waist circumference, 93.0 ± 91.7 ± 91.4 ± 10.7 0.766 cm 11.0 10.9 Waist/height, cm 0.54 ± 0.53 ± 0.54 ± 0.07 0.790 0.06 0.07 Conicity index 1.50 ± 1.30 ± 1.39 ± 0.62 0.269 0.76 0.35 Glucose, mg/dL 91.6 ± 9.2 93.6 ± 93.6 ± 9.6 0.517 11.6 Cholesterol, mg/dL Total 196 ± 30.1 198 ± 196 ± 31.6 0.937 32.7 LDL 124 ± 26.4 125 ± 127 ± 25.1 0.900 31.5 HDL 50.9 ±13.4 51.0 ± 49.8 ± 13.0 0.889 14.7 Triglycerides*, mg/dL 82 (67- 85 (65- 81 (63- 0.624 118) 121) 116) Physical activity, AU 3.08 ± 3.12 ± 3.12 ± 1.26 0.986 0.06 1.38 Data are expressed as the mean ± standard deviation, or percentages. CD, control drink; OJ, orange juice; EOJ, enriched orange juice; SBP, systolic blood pressure; DBP, diastolic blood pressure; Pulse pressure = SBP-DBP; BMI, body mass index; LDL, low- density lipoproteins; HDL, high-density lipoproteins * median (25th -75th percentiles). AU, arbitrary units: 0, inactive; 1, very low activity; 2, low activity; 3, moderately active; 4, very active. P for ANOVA with logarithmic transformation for triglycerides.

153

Table S3. Energy, nutrients, fibre and alcohol after 12 weeks of intervention. Variable Treatment P*

Control P OJ P Enriched OJ P

Energy, kcal/day

Baseline 2443 ± 638 2422 ± 618 2488 ± 627 12-week 2270 ± 691 0.033 2154 ± 610 0.001 2293 ± 565 0.019 NS HC, % energy

Baseline 37.5 ± 6.3 37.9 ± 5.8 37.5 ± 6.1 12-week 34.0 ± 8.4 0.008 35.8 ± 7.5 0.042 36.8 ± 7.9 0.533 NS HC, grams

Baseline 224 ± 55 227± 67 227 ± 66 12-week 187± 66 <0.001 190 ± 62 <0.001 205 ± 61 0.001 NS Protein, % energy

Baseline 17.6 ± 2.9 16.6 ± 2.4 17.6 ± 3.8

12-week 17.9 ± 3.6 0.578 17.7 ± 3.5 0.028 16.8 ± 2.9 0.142 0.031†

Protein, grams

Baseline 104 ± 29.5 99.0 ± 26.0 106 ± 28.9 12-week 100 ± 32.6 0.287 93.8 ± 28.6 0.204 94.3 ± 2.,3 0.003 NS Total fat, % energy

Baseline 43.0 ± 5.2 43.1 ± 5.7 41.9 ± 4.8

12-week 45.5 ± 7.7 0.036 43.8 ± 6.1 0.496 43.1 ± 5.2 0.230 NS

Total fat, grams

Baseline 120 ± 42.4 118 ± 33.8 118 ± 34.7 12-week 118 ± 44.8 0.759 107 ± 36.7 0.032 112 ± 32.6 0,310 NS

154

SFA, % energy

Baseline 12.1± 2.4 12.4 ± 2.8 12.0 ± 2.7

12-week 12.6 ± 3.1 0.233 13.3 ± 3.3 0.092 11.9 ± 2.3 0.886 NS

SFA, grams

Baseline 34.7± 14.8 34.2 ± 130 34.2 ± 12.0

12-week 33.8 ± 15.4 0.662 33.1 ± 14.2 0.572 30.8 ± 9.2 0.111 NS

Table S3 (cont.) MUFA ,% energy

Baseline 19.1 ± 4.1 19.9 ± 2.9 19.4 ± 4.3 12-week 20.1 ± 4.2 0.217 19.6 ± 3,8 0.566 19.5 ± 3.8 0.869 NS MUFA, grams

Baseline 53.3 ± 21.0 53.8 ± 15.5 53.0 ± 16.6

12-week 51.9 ± 21.2 0.641 46.5 ± 13.3 0.003 50.0 ± 14.3 0.270 NS PUFA, % energy

Baseline 8.3 ± 3.6 7.4 ± 2.5 7.6 ± 2.3

12-week 8.9 ± 2.8 0.140 7.6 ± 2.6 0.549 8.1 ± 2.9 0.320 NS

PUFA, grams

Baseline 22.8 ± 9.9 19.8 ± 7.6 21.4 ± 9.4 12-week 22.1 ± 9.2 0.612 18.9 ± 10.4 0.431 23.3 ± 16.9 0.495 NS

155

Fibre, g/daya

Baseline 23.1 (16.1-30.2) 21.0 (15.9-27.4) 24.0 (19.4-28.6) 12-week 18.5 (13.3-23-.7) 17.0 (13.3-21.2) < 0.001 19.0 (15.6-27.2) 0.028 NS <0.001

Alcohol, g/daya

Baseline 1.69 (0.00-8.8) 4.51 (0,56-10.7) 5.97 (0.36-22.8)

12-week 3.60 (0.02-13.6) 0.068 6.43 (0.66-13.1) 0.279 6.30 (0.00-14.9) 0.589 NS

Data are expressed as the mean ± standard deviation: a median (25-75th percentile). H C, carbohydrates; SFA, saturated fatty acids; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids. Intra-treatment comparisons by Student’s t test and Wilcoxon test for related samples. * P value for differences among treatments. ANOVA and Mann-Whitney tests. †P for Enriched OJ versus OJ.

156

Online Supplemental Figures for the manuscript

157

158

159

160

Project 3: Effect of the consumption of hesperidin in orange juice on the transcriptomic profile of subjects with pre- and stage 1 hypertension: a randomized controlled trial (CITRUS study) (Editor submitted)

161

162

Effect of the consumption of hesperidin in orange juice on the transcriptomic profile of subjects with pre- and stage 1 hypertension: a randomized controlled trial (CITRUS study)

Laura Pla-Pagà1,2, Rosa M Valls2,1*, Anna Pedret1,2*, Lorena Calderón- Pérez1,2, Elisabet Llauradó2, Judit Companys1,2, Cristina Domenech- Coca1, Nuria Canela3, Josep M del Bas1, Antoni Caimari1, Francesc Puiggròs3, María-Isabel Covas4, Lluís Arola3, Rosa Solà2,5

1Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain. 2Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut, Functional Nutrition, Oxidation and Cardiovascular Disease Group (NFOC-Salut), Reus, Spain. 3Eurecat, Centre Tecnològic de Catalunya, Centre for Omic Sciences, Joint Unit Universitat Rovira I Virgili-EURECAT, Reus, Spain. 4Nutritional Projects Assessment (Nuproas.es); Barcelona, Spain. 5Hospital Universitari Sant Joan de Reus, Reus, Spain.

*Corresponding authors: Rosa M. Valls Zamora, PhD. Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut, Functional Nutrition, Oxidation and Cardiovascular Disease Group (NFOC-Salut). C/Sant Llorenç 21, 43201 Reus, Spain. E-mail: [email protected] Anna Pedret Figuerola, PhD. Eurecat, Centre Tecnològic de Catalunya. Av/ Universitat 1, 43204 Reus, Spain. E-mail: [email protected]

163

Abbreviations CD: control drink CVD: cardiovascular diseases CVDR: cardiovascular diseases risk CVDRFs: cardiovascular diseases risk factors DBP: diastolic blood pressure EOJ: hesperidin-enriched orange juice IPA: ingenuity pathway analysis LncRNA: long non-coding RNA OJ: orange juice PBMCs: peripheral blood mononuclear cells SBP: systolic blood pressure

Keywords: transcriptomic analysis, hypertensive subjects, hesperidin, orange juice, PBMCs, plasma.

164

ABSTRACT

Scope Hesperidin exerts cardiovascular beneficial effects, but its mechanisms of action remain undefined. The aim of this work was to determine whether the sustained and single dose consumption of hesperidin in orange juice (OJ) and hesperidin-enriched orange juice (EOJ) can change the transcriptomic profile of subjects with pre- and stage 1 hypertension to elucidate possible mechanisms of action of the hesperidin. Methods and results For transcriptomic analysis, peripheral blood mononuclear cells were obtained from 37 subjects with pre- and stage 1 hypertension from CITRUS study who were randomized to receive for 12 weeks: control drink (CD; n=11), OJ (containing 392 mg of hesperidin; n=15) or EOJ (containing 670 mg of hesperidin; n=11). At baseline, a single dose 6- hour study in each group was also performed. After the single dose consumption, EOJ versus OJ, downregulated DHRS9 gene which is related with insulin resistance. Compared to CD, sustained consumption of EOJ downregulated 6 proinflammatory while after OJ consumption only 1 proinflammatory gene was downregulated. Moreover, sustained consumption of EOJ versus OJ, downregulated acute coronary syndrome gene related (SELENBP1). Conclusion A single dose consumption of EOJ could protect from insulin resistance. Moreover, EOJ decrease the expression of proinflammatory genes after sustained consumption providing a possible mechanism of action on inflammation pathway.

165

1. INTRODUCTION Hesperidin is a naturally occurring flavonoid present in citrus fruits and is found at particularly high concentrations in orange fruits and orange juice (OJ).(9) The beneficial effects of hesperidin on cardiovascular risk factors have been elucidated and integrated in a recently published systematic review of animal studies and randomized human clinical trials.(130) The beneficial effects of hesperidin consumption on the lipid profile and glucose levels have been observed in rats, but the results from human studies remain unclear.(130) Some studies have related hesperidin to hypotensive,(98,131) hypolipemiant,(132) anti- inflammatory(131) and antioxidant(99) effects. Consistently, we recently reported that the sustained and acute consumption of hesperidin in OJ decreases the systolic blood pressure (SBP) and pulse pressures in subjects with pre- and stage 1 hypertension, and the sustained consumption of hesperidin-enriched OJ (EOJ) enhances the postprandial effect of hesperidin compared with that obtained with a single dose.(133) Thus, the consumption of hesperidin via OJ could be an interesting strategy for the SBP levels and treatment of cardiovascular disease risk factors (CVDRFs).(134)

In contrast, the impacts of different bioactive compounds on gene expression have been clearly established and have gained much interest in research because better comprehension of these effects can be used to prevent, detect and treat chronic diseases.(135) Therefore, knowledge about nutrient-gene interactions is key for obtaining more information about new mechanisms of action, and transcriptomics can provide such knowledge because compares the transcriptomes, which are sets of RNA transcripts of cells, tissues or organisms under specific conditions. For complex diseases, such as cardiovascular disease

166

(CVD) and cancer, a transcriptomic approach offers the possibility to identify novel mechanisms and thus further characterize disease pathophysiology.

Different families of phenolic compounds have the ability to modify gene expression profiles.(53) Specifically, phenolic compounds from virgin olive oil improved cholesterol efflux gene expression in humans(12) and can modify the whole transcriptome to exert beneficial effects on CVD and cancer.(136) Similarly, other phenolic compounds such as flavonoids, such as resveratrol and hesperidin, have been linked to effects on the transcriptome. Thus, previous studies have shown that resveratrol consumption improves vascular function in older adults.(137) Moreover, the chronic consumption of hesperidin can change the expression of leucocyte genes to exert an anti-atherogenic and anti-inflammatory effects in overweight subjects.(138) However, the molecular mechanism through which hesperidin influences metabolic pathways via transcriptome changes remains unclear. To the best of our knowledge, a transcriptomic analysis that simultaneously assesses the impacts of sustained and acute hesperidin consumption in human peripheral blood mononuclear cells (PBMCs) of subjects with hypertension has not been performed. PBMCs are highly useful for demonstrating the capacity of cell systems to interact with nutrients and bioactive food compounds.(139,140)

The aim of this work was to determine whether the sustained and single dose consumption of hesperidin in OJ and EOJ can change the transcriptomic profile of PBMCs of subjects with pre- and stage 1 hypertension to elucidate possible mechanisms of action of the hesperidin and their role in CVD.

167

2. EXPERIMENTAL SECTION 2.1 Subjects The study population was included in our previous randomized clinical trial.(133) The subjects were 159 men and women with pre and stage 1 hypertension who did not smoke, had no family history of CVD, SBP ≥ 120 mm Hg and < 159 mm Hg, diastolic blood pressure (DBP) < 100 mm Hg,(34) and were not taking any anti-hypertensive medications.

2.2 Test drinks The three interventional drinks were provided by the Florida Department of Citrus from the USA: a control drink (CD), OJ with the natural content of hesperidin, and EOJ, which contains the 2S form of hesperidin, the naturally form in OJ and a bioavailable form (micronized 2S hesperidin; Ferrer Health Tech, Murcia, Spain).(103) The composition of the three test drinks is detailed in Supporting Information Table S1.

2.3 Dosage information The subjects orally consumed 500 mL/day of the corresponding test drink: CD (0 mg/day of hesperidin), OJ (345 mg/day of hesperidin) and EOJ (600 mg/day of hesperidin).

2.4 Study design A randomized, parallel, double-blind, placebo-controlled clinical trial was performed and registered at Clinical-Trials.gov under NCT02479568. All volunteers were randomly assigned to one of the three intervention groups: CD, OJ and EOJ groups. The sustained intervention was conducted for 12 weeks and nested, two 6-hour single dose studies, one at baseline and the other after 12 weeks, were performed.(133) For the transcriptomic analysis in the sustained study,

168

PBMCs were obtained from blood samples collected under fasting conditions at baseline and after 12 weeks. For the transcriptomic analysis in the single dose study, PBMCs were obtained at baseline and after 6 hours only in the single dose study conducted at the beginning of the study.

2.5 Transcriptomic analysis Whole gene expression in PBMCs from whole blood samples collected under fasting conditions in both the dose-response and sustained studies was assessed using the Agilent Microarray Platform (Agilent Technologies, Santa Clara, California, USA). Total PBMC RNA was isolated by Ficoll gradient separation (GE Healthcare Bio Sciences, Barcelona, Spain). The RNA yield was quantified with a Nanodrop UV- VIS spectrophotometer, and the RNA integrity was measured with an Agilent 2100 Bioanalyzer using the Total RNA Nano kit and the protocol Eukaryote Total RNA Nano (Agilent Technologies, Santa Clara, California, USA). Total RNA (100 ng) from the PBMCs was labelled with Cy3 (ref: 5190-2305, Agilent Technologies, Santa Clara, California, USA) and hybridized using a Gene Expression Hybridization Kit (ref: 5188-5242, Agilent Technologies, Santa Clara, California, USA), on an Agilent SurePrint G3 Human Gene Expression 8x60K v3 Microarray (Design ID 072363) with SurePrint Technology (Agilent Technologies, Ref. G4851C). Image scanning was performed with a 3 μm resolution using an Agilent Microarray Scanner System with SureScan High-Resolution Technology (ref: G2565CA, Agilent Technologies, Santa Clara, California, USA). The Feature Extraction version 12.0.1.1 software (Agilent Technologies) was used for data extraction.

169

Gene expression and real-time PCR analysis To analyse the gene expression in the samples and validate the DNA array results, cDNA was synthesized using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4 Barcelona, Spain). A MyGene Series Peltier Thermal Cycler (LongGene Scientific, Zhejiang, China) was used for reverse transcription. The reaction was performed according to the manufacturer’s instructions. The cDNA was amplified by quantitative reverse transcriptase polymerase chain reaction using LightCycler 480 SYBR Green I Master (Roche Diagnostic, Sant Cugat del Vallès, Barcelona, Spain) and a LightCycler 480 II system (Roche Diagnostic, Sant Cugat del Vallès, Barcelona, Spain). The primers used for the different genes are described in Supporting Information Table S2 and were obtained from Biomers.net (Ulm, Germany). The fold changes in the mRNA levels were calculated as percentages compared with the control group (CD) using the -2∆∆Ct method (141) and the RPLP0 gene was used as an endogenous control. Real-time PCR analysis was performed in a subset of six genes: DHRS9, DSP, TNF, NAMPT, PTX3 and IER3; and in a subsample of the samples collected: 22 samples for DHRS9, 18 for DSP, 24 for TNF, 19 for NAMPT, 17 for PTX3 and 20 for IER3.

2.7 Data analysis Quality control was performed through principal component analyses. Statistical comparisons were performed by Student’s t-test or Welch’s t-test. Differentially expressed genes were genes with that met the following criteria: P-value < 0.05 and fold change <= -0.58 or >= 0.58 in the log2 scale (corresponding to a 1.5-fold difference in natural scale). Treatment comparisons were performed with an ANCOVA model adjusted for age, sex and baseline values. Statistical significance

170 was defined by a P value ≤ 0.05 from a two-sided test. SPSS for Windows (version 21; IBM Corp., Armonk, New York, USA) was used for the analyses. Differentially expressed genes were subjected to functional and biochemical pathway analyses using Gene Ontology and the KEGG (Kyoto Encyclopedia of Genes and Genomes (www.genome.jp/kegg) and PANTHER (protein annotation through evolutionary relationship classification system (http://www.pantherdb.org/) biochemical pathway databases, respectively. The analysis was performed using GeneCodis (http://genecodis.dacya.ucm.es) software. For biological interpretation, the functions, pathways and networks of the results that showed significant differences between the groups were analysed by Ingenuity pathway analysis (IPA; www.ingenuity.com), which explores the possible metabolic cell signalling pathways that were up- and downregulated after the intervention.

3. RESULTS 3.1 Baseline characteristics of the participants PBMC samples were collected from 37 of the 159 subjects included in the original randomized clinical trial after both the single dose and sustained interventions, and these subjects were thus candidates for transcriptomic profile evaluation.(133) These 37 participants were assigned to the three groups: CD, N = 11; OJ, N = 15; and EOJ, N = 11. The baseline characteristics of the participants of each group are detailed in Table 1. The volunteers ranged in age from 41 to 65 years, had a normal weight or were overweight, and had pre- and stage 1 hypertension. No differences in the baseline characteristics were found among the three groups.

171

3.2 Clinical results In accordance with our previously published results,(133) the consumption of hesperidin in OJ and EOJ for 12 weeks decreased the SBP of subjects with pre- and stage 1 hypertension by an average of - 6.35 mm Hg and -7.36 mm Hg, respectively. Additionally, the pulse pressure, which is a recognized independent risk factor for CVD, decreased by -2.41 mm Hg after the consumption of EOJ. No significant changes in the DBP were observed. Additionally, the homocysteine and uric acid levels decreased after the consumption of OJ and EOJ. In contrast, the acute consumption at baseline of a single dose of hesperidin in EOJ reduced the SBP, pulse pressure and homocysteine levels, and after 12 weeks of treatment resulted in greater changes in these variables as well as a decrease in the DBP. Moreover, as a new result, the plasminogen activator inhibitor type 1 (PAI-1) levels decreased after EOJ consumption versus CD (Mean change: -0.191; 95% CI: -0.32; -0.06; P=0.004) and also versus OJ (Mean change: -0.199; 95% CI: -0.33; -0.07; P=0.002).

3.3 Gene expression profile After sustained and acute consumption, the number of statistically or borderline differentially expressed genes (P < 0.05 or P < 0.10, respectively) after OJ and EOJ compared with the levels obtained with the CD are represented in Figure 1. Compared with their expression in the CD group, 3 genes were commonly differentially or potentially differentially expressed after consumption of a single dose of hesperidin in OJ and EOJ. In addition, compared with their expression in the CD group, after the consumption of hesperidin in OJ and EOJ for 12 weeks, 16 genes were commonly differentially or potentially differentially expressed compared with their levels in the CD group.

172

Gene expression after dose-response intervention Comparisons of the single dose gene expression changes in PBMCs of individuals with pre-a and stage 1 hypertension among the three treatment groups are shown in Table 2. Three differentially expressed genes (P < 0.05) were observed after the single dose consumption of OJ compared with the CD. Two of these three genes, DHRS9 and PKDL1, were significantly upregulated, while the other gene, TNFAIP3, was downregulated. Moreover, 5 genes were borderline differentially expressed (P < 0.10) after OJ consumption. The comparison of the EOJ and CD groups revealed no differentially expressed genes, but 6 genes were borderline differentially expressed (P < 0.10). Finally, the comparison of the EOJ with OJ group revealed that DHRS9 gene was significantly downregulated (P < 0.05). All these genes are mainly related to the inflammation pathways and insulin resistance.

Gene expression after sustained intervention Comparisons of the gene expression changes in PBMCs from individuals with pre- and stage 1 hypertension among the three treatment groups after 12 weeks of sustained consumption of hesperidin are shown in Table 3. After 12 weeks, we identified 12 genes that were differentially expressed in the OJ group compared with the CD group, and these included seven upregulated genes (CCL20, FAM53B, LINC 01220, lncRNA SNRPD3-2, lncRNA NFKBID-1, lncRNA PDE3B-1, and LOC101929524) and five downregulated genes (TNF, TMPO- AS1, BPIFB3, lncRNA ACOT-13 and CCT8-1). Moreover, 13 genes were borderline differentially expressed (P < 0.10) after OJ consumption. After 12 weeks, 18 genes showed differential expression in the EOJ group compared with the CD group, and these included four upregulated genes (DSP, FAM53B, lncRNA SNRPD3-2 and lncRNA

173

SLC39A8-1) and 14 downregulated genes (TNF, IL1B, CCL3L3, CXCL2, CXCL8, PTGS2, IER3, PTX3, KMT22-AS1, ATP2B1-AS1, lncRNA CCT8-1, lncRNA GRK3-1, lncRNA CCDC117-1 and LOC644090). Moreover, 17 genes were borderline differentially expressed (P < 0.10) after EOJ consumption. Finally, after 12 weeks, 4 genes showed differential expression between the EOJ and OJ groups: one gene (lncRNA IYD) was upregulated, and 3 genes (SELENBP1, ALAS2 and BTBD19) were downregulated. The 12 differentially expressed genes after OJ were related to inflammation (2 genes), cancer (1 gene) and unknown functions (9 genes). After EOJ, the 18 differentially expressed genes were related to inflammation (6 genes), endothelial function (2 genes), hypertension (1 gene), and unknown functions (9 genes). A summarized table of results are detailed in Supporting Information Table S4.

Pathway analysis IPAs were performed only with the genes that showed significantly and borderline significantly differential expression after sustained consumption of hesperidin because the number of differentially expressed genes after acute consumption was insufficient for the pathway analysis. None of the long noncoding RNAs (lncRNAs) that were showed significantly or borderline significantly differential expression after 12 weeks of hesperidin consumption could be used for the pathway-based IPA due to the scarcity of the related literature. The biological networks based on the differentially expressed genes after sustained consumption of OJ and EOJ are represented in Figure 2. The overlapping graphical representation of the two most important

174 networks identified by the IPAs of the differentially expressed genes after OJ and EOJ consumption compared to CD consumption, are shown in panels A and B, respectively, and the up- and downregulated genes are shown in red and green colours, respectively. The top canonical pathways, diseases and functions regulated by sustained consumption of OJ and EOJ and the implicated genes are shown in Table S3 in the Supplementary Information. The top diseases and functions related to the differentially expressed genes identified after the consumption of hesperidin in OJ for 12 weeks were the inflammatory response, cell-to-cell signalling and interaction, organismal injury and abnormalities, and the haematological and cardiovascular systems. After the consumption of hesperidin in EOJ for 12 weeks, the top diseases and functions were the inflammatory response, organismal injury and abnormalities, cellular growth and proliferation, lipid metabolism and the haematological system.

3.4 Results of the real-time PCR analysis Figure 3 shows the real-time PCR validation of the arrays results for a subset of 6 genes, which demonstrated that the directions of the differences induced by the interventions were consistent with both techniques.

4. DISCUSSION In the present work, we demonstrated that the consumption of 500 mL/day for 12 weeks of OJ and EOJ can change the transcriptomic profile of PBMCs of subjects with pre- and stage 1 hypertension. Our results showed the single dose consumption of EOJ compared to OJ, downregulated the gene DHRS9 reducing the insulin resistance.

175

Moreover, the sustained consumption of EOJ, which provides a higher dose of hesperidin, can induce greater changes in the expression profile of PBMCs, compared with those obtained with the lower dose of hesperidin in OJ, since EOJ downregulated more pro-inflammatory genes.

Interestingly, the differential expression of a gene related to insulin resistance was only observed after the single dose consumption of hesperidin in OJ and EOJ but not after sustained consumption for 12 weeks. The consumption of a single dose of hesperidin in OJ significantly upregulated the expression of DHRS9, whereas the consumption of a single dose of hesperidin in EOJ downregulated DHRS9 gene expression compared to CD. DHRS9, an oxidoreductase/decarboxylase, promotes insulin resistance by activating lipid metabolism.(142) Accordingly, we hypothesized that the presence of fructose in OJ could increase DHRS9 expression and thereby increases fructose metabolism, the plasma triglyceride levels and insulin resistance;(142) whereas the higher doses of hesperidin in EOJ could counteract the effects of the fructose content to induce downregulation of the DHRS9 gene. The evidence suggests that the consumption of 50-100 mg/kg body weight/day hesperidin improves insulin resistance and decreases the glucose plasma levels in rats.(143,144) Because this dose range of hesperidin for animals is equivalent to the hesperidin dose present in EOJ (670 mg/day), the results might explain the different gene expression profiles detected after a single dose between the two test drinks, whereas positive results were observed after EOJ consumption.

176

Moreover, the consumption of a single dose of hesperidin in EOJ induced the borderline significant upregulation of different pro- inflammatory genes, including the gene encoding the chemokine CXCL2. CXCL2, which is also called macrophage inflammatory protein 2-alpha, produces endogenous signals during the first steps of inflammasome activation to generate pro-inflammatory cytokines(145) A recently published review reported that CXCL2 overexpression contributes to atherosclerotic plaque formation, inflammation in obesity and the induction of diabetes, which demonstrates that CXCL2 might be a therapeutic target in CVD.(146) Additionally, the consumption of a single dose of EOJ borderline significantly downregulated the expression of the CCL3-encoding gene. CCL3 is a chemokine with pro- inflammatory activity and serves as an independent risk factor in subjects with acute coronary syndromes because high levels of CCL3 are associated with short-term mortality.(147) Additionally, high levels of this chemokine are associated with future ischaemia events in subjects with acute myocardial infarction (148) Thus, downregulated CCL3 gene expression can suggest a role for this chemokine in the prevention of inflammation related to cardiovascular disease risk (CVDR).

The sustained consumption of hesperidin in OJ and EOJ induces significantly differential expression of different pro-inflammatory genes. In this sense, the consumption of OJ and EOJ for 12 weeks significantly downregulated TNF gene expression. Experimental studies have shown that the cytokine TNF is associated with hypertension and renal injuries in hypertensive rats and mice.(149) In addition, the intracerebroventricular administration of TNF to

177 normotensive and hypertensive rats increases their blood pressure and thereby influences angiotensin type 1 receptors.(149) Therefore, the downregulation of TNF might be beneficial for the improvement of high blood pressure levels and thereby influences the angiotensin II levels and oxidative stress in hypertension.(150) If these effects can be translated to humans, they can at least partly explain the decrease in SBP observed in subjects with pre- and stage 1 hypertension.(133) IL1B was also significantly downregulated after 12 weeks of EOJ compared with CD consumption. IL1B is a cytokine with cogent links to atherosclerosis and other inflammatory diseases.(151) Previous studies have shown that high levels of ILB1 promote atherothrombosis and thereby increase CVDR.(151) Additionally, increased levels of ILB1 suggest an important role of this cytokine in hypertension pathogenesis.(152) In addition, PTGS2 was borderline significantly downregulated after 12 weeks of EOJ consumption. PTGS2 (also named COX-2) induces pro-inflammatory cytokines and thereby stimulates cell proliferation and increases metastatic potential through inflammatory pathways.(153) PTGS2 is also an important enzyme in prostaglandin synthesis, and its overexpression increases the risk of ischaemic stroke.(154) In this sense, CITRUS study showed decreased levels of PAI-1 after EOJ consumption and reduced homocysteine levels after OJ and EOJ consumption for 12 weeks(133), and both results are related with an improvement of inflammation, that which observed also at transcriptome level in the present analysis.

In contrast, unlike our previously positive results regarding decreases in the expression of pro-inflammatory genes, some anti-inflammatory- related genes were significantly, or borderline significantly

178 downregulated after the single dose and sustained consumption of hesperidin in OJ and EOJ. Specifically, TNFAIP3, which is related to vascular disease, was significantly downregulated after a single dose of OJ compared with CD. The overexpression of the TNFAIP3 gene is related to suppression of adhesion protein expression at sites of atherosclerotic plaque formation; therefore, lower levels of TNFAIP3 are associated with an increased risk of atherosclerosis in subjects with type 2 diabetes.(155) An aspect that should be considered is that the decrease in the expression of anti-inflammatory genes could be explained by the fructose content of fruit juices, although natural and enriched juices were used in the present study. A high consumption of fructose can increase CVDR due to increases in inflammation pathways,(142) but some controversial studies have only showed an association between cardiometabolic risk factors and fructose content though the consumption of artificially sweetened beverages rather than fruit juices.(156) The unclear conclusions might be due to the presence of fibres and phenolic compounds, such as hesperidin, or other beneficial bioactive compounds, such as vitamins and minerals, in fruit juices, including OJ. Hence, more studies in this field are needed to obtain conclusive results regarding the effects of fruit juices and the fructose content on inflammation and CVD factors.

Moreover, our results showed that the gene selenium binding protein 1 (SELEBP1) was downregulated after EOJ consumption for 12 weeks. The elevated levels of SELENBP1 was related with a high risk of adverse cardiac events and death(157). Thus, the downregulation of these gene could be beneficial since it is considered a biomarker for cardiac events. Additionally, the genes PTX3 and NAMPT previously

179 published in CITRUS study(133), were changed after 12 weeks of OJ and EOJ consumption, related with the improvement of blood pressure levels. Also, the genes DSP and IER3, which are related with an improvement of endothelial function, were differentially expressed after EOJ sustained consumption.

Moreover, our findings suggest that both the single dose and sustained consumption of hesperidin in OJ and EOJ for 12 weeks can change the expression of several lncRNAs. lncRNAs, which are RNA transcripts that do not encode proteins, have been implicated in numerous biological processes and diseases.(158) LNCipedia, a comprehensive compendium of human lncRNAs with an integrated database that offers the annotation of thousands of lncRNA transcript sequences, was searched for the differentially expressed lncRNAs, including both up- and downregulated lncRNAs, after the consumption of OJ and EOJ compared with the CD. However, although some of these lncRNAs were identified in previous studies, most of the identified lncRNAs are unknown, and their functions and involvement in diseases have not yet been assessed. Accordingly, further research is needed in this topic because some studies have shown that lncRNAs can act as key regulators of the inflammatory response(159) and can play an important role in the cardiovascular system.(160)

One strength of the present work is that the study constitutes the first evaluation of the transcriptomic profile of human PBMCs after the consumption of hesperidin in OJ and EOJ. The assessment of gene expression in human PBMCs can demonstrate their capacity to reflect the distinct gene expression signatures of several diseases linked to

180 cardiovascular health(140,161) and the gene expression signatures after interaction with bioactive food compounds, such as flavonoids. In contrast, this work provides the first evaluation of the transcriptomic profile of human PBMCs after the acute consumption of hesperidin in OJ and EOJ. The evidence demonstrates the capacity of phenolic compounds to exert postprandial effects on several systems, and the accumulation of acute functional changes can induce chronic physiological alterations,(162) and thus has a positive impact on the target systems. The insights on postprandial actions can offer the possibility to study unknown mechanisms for the beneficial effects of hesperidin in human health. However, whether the changes in the transcriptomic profile of PBMCs are maintained over time and whether the observed changes might improve long-term CVDRFs remain unclear. Therefore, larger clinical trials and trials that include patients with hypertension and other CVDRFs are needed to better understand these clinical findings. Moreover, in the present work, the transcriptomic analysis of the single dose intervention was only performed at the beginning of the clinical trial. Thus, a second PBMC analysis in the second single dose study conducted at 12 weeks could provide evidence regarding the effects of chronic phenolic compounds intake on the possible optimization of the acute intake observed in previous studies.(133)

In conclusion, the single dose and sustained consumption of hesperidin in OJ and EOJ changed the transcriptome of PBMCs of subjects with pre- and stage 1 hypertension. The single dose consumption of higher doses of hesperidin could induce a better response than the consumption of the naturally occurring doses of hesperidin in OJ because of their

181 improvement of insulin resistance. Moreover, the sustained consumption of hesperidin in EOJ decrease the expression of proinflammatory genes providing a possible mechanism of action on inflammation pathway and thereby could induces beneficial effects on the cardiovascular system. However, some results are unclear and controversial; therefore, more studies are needed to confirm and clarify the biological effects of hesperidin consumption on the transcriptomic profile.

5. REFERENCES

[1] D. Del Rio, A. Rodriguez-Mateos, J. P. E. Spencer, M. Tognolini, G. Borges, A. Crozier, Antioxid. Redox Signal. 2013, DOI 10.1089/ars.2012.4581. [2] L. Pla-Paga, J. Companys, L. Calderon-Perez, E. Llaurado, R. Sola, R. M. Valls, A. Pedret, Nutr. Rev. 2019, 77, 1–20. [3] F. Homayouni, F. Haidari, M. Hedayati, M. Zakerkish, K. Ahmadi, Phyther. Res. 2018, 32, 1073–1079. [4] C. Wunpathe, P. Potue, P. Maneesai, S. Bunbupha, P. Prachaney, U. Kukongviriyapan, V. Kukongviriyapan, P. Pakdeechote, Am. J. Chin. Med. 2018, 46, 1–17. [5] J. Shen, K. A. Wilmot, N. Ghasemzadeh, D. L. Molloy, G. Burkman, G. Mekonnen, M. C. Gongora, A. A. Quyyumi, L. S. Sperling, Annu. Rev. Nutr. 2015, DOI 10.1146/annurev-nutr- 011215-025104. [6] P. Maneesai, S. Bunbupha, P. Potue, T. Berkban, U. Kukongviriyapan, V. Kukongviriyapan, P. Prachaney, P. Pakdeechote, Nutrients 2018, 10, 1–15. [7] R. M. Valls, A. Pedret, L. Calderón-Pérez, E. Llauradó, L. Pla-

182

Pagà, J. Companys, A. Moragas, F. Martín-Luján, Y. Ortega, M. Giralt, M. Romeu, L. Rubió, J. Mayneris-Perxachs, N. Canela, F. Puiggrós, A. Caimari, J. M. Del Bas, L. Arola, R. Solà, Eur. J. Nutr. 2020, DOI 10.1007/s00394-020-02279-0. [8] A. Ahmadi, A. Shadboorestan, Nutr. Cancer 2016, 68, 29–39. [9] L. Bordoni, R. Gabbianelli, Biochimie 2019, 160, 156–171. [10] V. Tufarelli, E. Casalino, A. G. D’Alessandro, V. Laudadio, Curr. Drug Metab. 2017, DOI 10.2174/1389200218666170925124004. [11] M. Farràs, S. Arranz, S. Carrión, I. Subirana, D. Muñoz- Aguayo, G. Blanchart, M. Kool, R. Solà, M. J. Motilva, J. C. Escolà-Gil, L. Rubió, S. Fernández-Castillejo, A. Pedret, R. Estruch, M. I. Covas, M. Fitó, Á. Hernáez, O. Castañer, Nutrients 2019, 11, 1–10. [12] S. D’Amore, M. Vacca, M. Cariello, G. Graziano, A. D’Orazio, R. Salvia, R. C. Sasso, C. Sabbà, G. Palasciano, A. Moschetta, Biochim. Biophys. Acta - Mol. Cell Biol. Lipids 2016, 1861, 1671–1680. [13] R. M. Pollack, N. Barzilai, V. Anghel, A. S. Kulkarni, A. Golden, P. O’Broin, D. A. Sinclair, M. S. Bonkowski, A. J. Coleville, D. Powell, S. Kim, R. Moaddel, D. Stein, K. Zhang, M. Hawkins, J. P. Crandall, Journals Gerontol. - Ser. A Biol. Sci. Med. Sci. 2017, 72, 1703–1709. [14] D. Milenkovic, C. Deval, C. Dubray, A. Mazur, C. Morand, PLoS One 2011, 6, 1–9. [15] C. Cappuzzello, M. Napolitano, D. Arcelli, G. Melillo, R. Melchionna, L. Di Vito, D. Carlini, L. Silvestri, S. Brugaletta, G. Liuzzo, F. Crea, M. C. Capogrossi, Physiol. Genomics 2009,

183

38, 233–240. [16] S. D’Amore, M. Vacca, G. Graziano, A. D’Orazio, M. Cariello, N. Martelli, G. Di Tullio, R. Salvia, G. Grandaliano, A. Belfiore, F. Pellegrini, G. Palasciano, A. Moschetta, Biochim. Biophys. Acta - Mol. Basis Dis. 2013, 1832, 2289–2301. [17] M. F. Piepoli, A. W. Hoes, S. Agewall, C. Albus, C. Brotons, A. L. Catapano, M. T. Cooney, U. Corrà, B. Cosyns, C. Deaton, I. Graham, M. S. Hall, F. D. R. Hobbs, M. L. Løchen, H. Löllgen, P. Marques-Vidal, J. Perk, E. Prescott, J. Redon, D. J. Richter, N. Sattar, Y. Smulders, M. Tiberi, H. B. Van Der Worp, I. Van Dis, W. M. M. Verschuren, S. Binno, G. De Backer, M. Roffi, V. Aboyans, N. Bachl, S. Carerj, L. Cho, J. Cox, J. De Sutter, G. Egidi, M. Fisher, D. Fitzsimons, O. H. Franco, M. Guenoun, C. Jennings, B. Jug, P. Kirchhof, K. Kotseva, G. Y. H. Lip, F. Mach, G. Mancia, F. M. Bermudo, A. Mezzani, A. Niessner, P. Ponikowski, B. Rauch, A. Stauder, G. Turc, O. Wiklund, S. Windecker, J. L. Zamorano, S. Achenbach, L. Badimon, G. Barón-Esquivias, H. Baumgartner, J. J. Bax, V. Dean, Ç. Erol, O. Gaemperli, P. Kolh, P. Lancellotti, P. Nihoyannopoulos, A. Torbicki, A. V. Carneiro, B. Metzler, R. Najafov, V. Stelmashok, C. De Maeyer, M. Dilić, I. Gruev, D. Miličić, H. Vaverkova, I. Gustafsson, I. Attia, D. Duishvili, J. Ferrières, N. Kostova, Z. Klimiashvili, R. Hambrecht, K. Tsioufis, E. Szabados, K. Andersen, C. Vaughan, B. Zafrir, S. Novo, K. Davletov, F. Jashari, A. Kerimkulova, I. Mintale, G. Saade, Z. Petrulioniene, C. Delagardelle, C. J. Magri, V. Rudi, L. Oukerraj, B. E. Çölkesen, H. Schirmer, R. P. Dos Reis, D. Gherasim, S. Nedogoda, M. Zavatta, V. Giga, S. Filipova, L. R.

184

Padial, A. Kiessling, A. Mahdhaoui, D. Ural, E. Nesukay, C. Gale, Eur. Heart J. 2016, 37, 2315–2381. [18] B. N. Salden, F. J. Troost, E. de Groot, Y. R. Stevens, M. Garcés-Rimón, S. Possemiers, B. Winkens, A. A. Masclee, Am. J. Clin. Nutr. 2016, 104, 1523–1533. [19] K. J. Livak, T. D. Schmittgen, Methods 2001, 25, 402–408. [20] M. Boro, K. N. Balaji, J. Immunol. 2017, 199, 1660–1671. [21] L. Y. Guo, F. Yang, L. J. Peng, Y. B. Li, A. P. Wang, Clin. Exp. Hypertens. 2019, 00, 1–10. [22] S. C. A. de Jager, B. W. C. Bongaerts, M. Weber, A. O. Kraaijeveld, M. Rousch, S. Dimmeler, M. P. van Dieijen-Visser, K. B. J. M. Cleutjens, P. J. Nelemans, T. J. C. van Berkel, E. A. L. Biessen, PLoS One 2012, 7, DOI 10.1371/journal.pone.0045804. [23] S. C. A. de Jager, A. O. Kraaijeveld, R. W. Grauss, W. de Jager, S. S. Liem, B. L. van der Hoeven, B. J. Prakken, H. Putter, T. J. C. van Berkel, D. E. Atsma, M. J. Schalij, J. W. Jukema, E. A. L. Biessen, J. Mol. Cell. Cardiol. 2008, 45, 446–452. [24] A. Segiet, P. Smykiewicz, P. Kwiatkowski, T. Żera, Cytokine 2019, 113, 185–194. [25] E. Mehaffey, D. S. A. Majid, Am. J. Physiol. - Ren. Physiol. 2017, 313, F1005–F1008. [26] P. Libby, J. Am. Coll. Cardiol. 2017, 70, 2278–2289. [27] J. Qi, X. F. Zhao, X. J. Yu, Q. Y. Yi, X. L. Shi, H. Tan, X. Y. Fan, H. L. Gao, L. Y. Yue, Z. P. Feng, Y. M. Kang, Cardiovasc. Toxicol. 2016, 16, 298–306. [28] S. J. Desai, B. Prickril, A. Rasooly, Nutr. Cancer 2018, 70, 350– 375.

185

[29] Z. Miao, M. Guo, S. Zhou, X. Sun, F. Wang, H. Lu, Z. Cui, Exp. Ther. Med. 2018, 16, 61–66. [30] R. Chen, J. Yan, P. Liu, Z. Wang, C. Wang, Metab. Brain Dis. 2017, 32, 667–673. [31] H. Lai, Q. Chen, X. Li, Y. Ma, R. Xu, H. Zhai, F. Liu, B. Chen, Y. Yang, Int. J. Clin. Exp. Med. 2015, 8, 21487–21496. [32] A. Seidi, S. Mirzaahmadi, K. Mahmoodi, M. Soleiman- Soltanpour, Mol. Biol. Res. Commun. 2018, 7, 17–24. [33] A. Ma, B. A. Malynn, Nat. Rev. Immunol. 2012, 12, 774–785. [34] Vasanti S Malik, F. B. Hu, J Am Coll Cardiol 2015, 66, 1615– 1624. [35] T. A. Khan, J. L. Sievenpiper, Eur. J. Nutr. 2016, 55, 25–43. [36] R. Sundaram, E. Nandhakumar, H. Haseena Banu, Toxicol. Mech. Methods 2019, 29, 644–653. [37] H. Xuguang, T. Aofei, L. Tao, Z. Longyan, B. Weijian, G. Jiao, Phyther. Res. 2019, 33, 1697–1705. [38] E. C. Kühn, A. Slagman, E. C. D. Kühn-Heid, J. Seelig, C. Schwiebert, W. B. Minich, C. Stoppe, M. Möckel, L. Schomburg, J. Trace Elem. Med. Biol. 2019, 52, 247–253. [39] X. Sun, M. S. S. Haider Ali, M. Moran, Biochem. J. 2017, 474, 2925–2935. [40] N. W. Mathy, X. M. Chen, J. Biol. Chem. 2017, 292, 12375– 12382. [41] C. Jiang, N. Ding, J. Li, X. Jin, L. Li, T. Pan, C. Huo, Y. Li, J. Xu, X. Li, Brief. Bioinform. 2019, 20, 1812–1825. [42] D. Esser, J. M. Geleijnse, J. C. Matualatupauw, J. I. Dower, D. Kromhout, P. C. H. Hollman, L. A. Afman, PLoS One 2018, 13, 1–15.

186

[43] N. Osakabe, J. Terao, Nutr. Rev. 2018, 76, 174–186.

ACKNOWLEDGEMENTS

Author contributions The human randomized clinical trial was conducted by LP-P, LCP, EL, JC, RMV, AP and RS. LP-P wrote the manuscript and RMV, AP and RS provided feedback on the manuscript. CD and LP-P realized the qRT-PCR analysis. All the authors have read the manuscript. We thank the Centre for Omics Sciences (COS) Joint Unit of the Universitat Rovira i Virgili-Eurecat, for their contribution to the processing of the samples and transcriptomic analysis. We thank Nutritional Projects Assessment (Nuproas.es) for their contribution to the statistical analysis.

Funding The Florida Department of Citrus, an executive agency of the state of Florida, USA, provided a grant and study drinks to the Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain. AP has Torres Quevedo contract (Subprograma Estatal de Incorporación, Plan Estatal de Investigación Cientíﬁca y Técnica y de Innovación).

Declaration of interest The authors declare that they have not conflicts of interest with the content of this article.

187

Table 1. Baseline characteristics of the participants by intervention group (N=37). Variable CD OJ EOJ P

(n=11) (n=15) (n=11)

Age 51.9 ± 12.8 41.0 ± 14.2 49.3 ± 5.9 0.061

Females, % 18.2 33.3 7.3 0.691

SBP, mm Hg 130.4 ± 9.8 134.9 ± 11.9 139.9 ± 11.8 0.156

DBP, mm Hg 74.6 ± 9.8 79.9 ± 9.2 82.1 ± 9.8 0.182

Pulse pressure, mm Hg 58.4 ± 9.7 55.0 ± 10.7 57.8 ± 7.2 0.623

Weight, kg 75.6 ± 10.0 76.9 ± 11.9 77.8 ± 11.8 0.905

BMI, kg/m2 25.5 ± 3.2 26.3 ± 3.5 26.3 ± 3.1 0.812

Waist circumference, 92.3 ± 9.1 91.8 ± 11.8 92.8 ± 7.7 0.966 cm

Glucose, mg/dL 95 ± 5.6 94 ± 9.3 97 ± 2.8 0.559

Cholesterol, mg/dL

Total 201 ± 42 187 ± 28 194 ± 19 0.507

LDL 134 ± 33 111 ± 27 129 ± 16 0.078

HDL 49.2 ± 11.9 54.6 ± 21.5 44.7 ± 13.9 0.347

Triglycerides, mg/dL 88 ± 20 104 ± 49 102 ± 39 0.594

Physical activity, UA 2.91 ± 1.7 3.33 ± 1.2 3.63 ± 0.5 0.398

Data expressed as mean ± standard deviation, or percentages. Abbreviations: CD, control drink; OJ, orange juice; EOJ, hesperidin-enriched orange juice; SBP, systolic blood pressure; DBP, diastolic blood pressure; BMI, body mass index; LDL, low density lipoprotein; HDL, high density lipoprotein; UA, arbitrary units: 0, inactive. 1, very low activity. 2, low activity. 3, moderately active. 4, very active.

188

Table 2. Comparisons of dose-response gene expression changes among treatments in PBMCs of mildly hypertensive individuals.

Gene Gene Name Post-int values mean (SE) Changes among treatments

Symbol Control OJ EOJ OJ vs Control EOJ vs Control EOJ vs OJ

(n=11) (n=15) (n=11) Mean P Mean P Mean P

(95%CI) (95%CI) (95%CI)

Inflammation related genes

Anti-inflammatory

NFKB1A Nuclear factor of kappa 7.19 6.23 6.23 -0.968 0.066 -0.962 0.064 0.006 0.991

light polypeptide gene (0.37) (0.32) (0.35) (-2.0;0.07) (-2.0;0.06) (-0.98;1.0)

enhancer in B-cells

inhibitor, alpha

TNFAIP3 Tumor necrosis factor, 4.00 2.82 3.11 -1.171 0.022 -0.889 0.073 0.282 0.549

alpha-induced protein 3 (0.35) (0.31) (0.34) (-2.2; -0.18) (-1.9;0.09 (-0.67;1.2)

194

TNFAIP6 Tumor necrosis factor, -2.45 -2.80 -2.92 -0.356 0.357 -0.471 0.219 -0.115 0.755

alpha-induced protein 6 (0.27) (0.24) (0.26) (-1.1;0.42) (-1.2;0.29) (-0.86;0.63)

DUSP2 Dual specify 1.29 0.86 0.92 -0.422 0.138 -0.364 0.192 0.058 0.827

phosphatase 2 (0.20) (0.27) (0.19) (-0.99;0.14) (-0.92;0.19) (-0.48;0.60)

TGM2 Transglutaminase 2 (C -3.70 -3.55 -3.56 0.144 0.140 0.134 0.161 -0.009 0.917

polypeptide, protein- (0.07) (0.06) (0.07) (-0.05;0.34) (-0.06;0.32) (-0.19;0.18)

glutamine-gamma-GT

TMIGD3 Transmembrane and -3.80 -3.70 -3.68 0.098 0.167 0.112 0.110 0.014 0.831

immunoglobulin (0.07) (0.04) (0.05) (-0.04; 0.24) (-0.03;0.25) (-0.12;0.15)

domain containing 3

Pro-inflammatory

CCL3 Chemokine (C-C 6.05 4.95 4.45 -1.094 0.133 -1.278 0.073 -0.204 0.765

motif) ligand 3 (0.53) (0.44) (0.49) (-2.5;0.35) (-2.7;0.13) (-1.6;1.2)

195

CCL3L3 Chemokine (C-C 2.72 1.39 1.22 -1.333 0.106 -1.502 0.066 -0.169 0.827

motif) ligand 3-like 3 (0.58) (0.50) (0.56) (-3.0;0.30) (-3.1;0.11) (-1.7;1.4)

CCL4L2 Chemokine (C-C 4.85 4.11 4.11 -0.739 0.074 -0.741 0.069 -0.002 0.996

motif) ligand 4-like 2 (0.89) (0.25) (0.28) (-1.5;0.07) (-1.5;0.06) (-0.78;0.78)

CXCL2 Chemokine (C-X-C 1.22 0.09 -0.34 -1.132 0.158 -1.566 0.051 -0.434 0.566

motif) ligand 2 (0.56) (0.49) (0.54) (-2.7;0.46) (-3.1;0.005) (-31.9;1.1)

EREG Epiregulin -1.81 -2.67 -2.66 -0.868 0.239 -0.853 0.240 0.015 0.691

(0.52) (0.45) (0.50) (-2.3; 0.60) (-2.3; 0.60) (-1.4;1.4)

CLEC5A C-type lectin domain -4.40 -3.94 -4.21 0.454 0.056 -0.191 0.409 -0.268 0.236

family 5, member A (0.17) (0.15) (0.16) (-0.01; 0.93 (-0.27;0.66) (-0.72;0.18)

FKBP5 FK506 binding protein 1.51 1.46 1.46 -0.059 0.465 -0.050 0.529 0.009 0.908

5 (0.06) (0.05) (0.05) (-0.22;0.10) (-0.21;0.11) (-0.15;0.16)

DDIT4 DNA-damage- 3.20 3.13 3.22 -0.069 0.547 0.022 0.847 0.091 0.409

inducible transcript 4 (0.08) (0.07) (0.08) (-0.30;0.16) (-0.21;0.25) (-0.19;0.84)

196

Cardiovascular related

FMN1 Formin 1 -4.37 -4.90 -4.84 -0.532 0.163 -0.470 0.209 0.062 0.864

(0.27) (0.23) (0.26) (-1.3;0.23) (-1.2;0.28) (-0.66;0.79)

Obesity related

AREG Amphiregulin -3.21 -3.52 -3.28 -0.318 0.425 -0.079 0.841 0.239 0.529

(0.28) (0.25) (0.27) (-1.1;0.48) (-0.87;0.71) (-0.53;1.0)

Insulin resistance

DHRS9 Dehydrogenase/reduct 0.031 0.323 0.038 0.292 0.019 0.057 0.992 -0.285 0.017

ase (SDR family) (0.08) (0.07) (0.08) (0.05;0.53) (-0.23;0.24) (-0.51;-0.05)

member 9, transcript

variant 3

Others (Unknown function)

SLED1 Proteogly can 3 0.190 -0.956 -0.769 -1.146 0.078 -0.958 0.132 0.188 0.757

pseudogene non- (0.45) (0.40) (0.44) (-2.4;0.14) (-2.2;0.31) (-1.0;1.4)

coding RNA

197

GOS2 G0 /G1 switch gene 2 1.95 0.638 0.699 -1.317 0.152 -1.256 0.165 0.061 0.944

(0.65) (0.56) (0.62) (-3.1;0.51) (-3.0;0.54) (-1.7;1.8)

PKDL1 Polycystic kidney -2.51 -1.96 -2.24 0.549 0.047 0.263 0.324 -0.286 0.269

disease2 -like 1 (0.19) (0.17) (0.18) (-0.007; 1.1) (-0.27;0.80) (-0.80;0.23)

LncRNA Long non coding RNA -3.89 -3.67 -3.88 0.219 0.368 0.012 0.961 -0.207 0.372

CEP44-1 Centrosomal Protein (0.17) (0.15) (0.17) (-0.27;0.71) (-0.47;0.49) (-0.67;0.26

44-1

LncRNA Long non coding RNA -2.77 -3.01 -2.96 0.231 0.062 -0.188 0.121 0.043 0.709

ATOH8- Atonal homolog 8-2 (0.09) (0.07) (0.08) (-0.47;0.01) (-0.43;0.05) (-0.19;0.28)

LncRNA Long non coding RNA -4.05 -4.04 -4.06 0.017 0.905 -0.005 0.972 -0.021 0.872

ERP44-3 endoplasmic reticulum (0.10) (0.09) (0.10) (-0.26;0.30) (-0.28;0.27) (-0.29;0.25)

protein 44-3

198

XLOC- Long non coding -4.69 -4.87 -4.77 -0.184 0.535 -0.084 0.772 0.100 0.725 l2_01322 XLOC-l2-0132293 (0.21) (0.18) (0.20) (-0.78;0.41) (-0.67;0.50) (-0.47;0.67)

Data expressed as mean ± standard error (ES) or mean (95% confidence interval, CI). Inter treatment comparisons by ANCOVA Model adjusted by sex and age. Analyses performed with the IBM SPSS Statistics 21 Package. Abbreviations: OJ, orange juice; EOJ, hesperidin-enriched orange juice.

199

Table 3. Comparisons of gene expression changes among treatments after 12 weeks of sustained consumption in PBMCs of mildly hypertensive individuals.

Gene Symbol Gene Name Post-intervention values Changes among treatments

mean (SE)

Control OJ EOJ OJ vs Control EOJ vs Control EOJ vs OJ

(n=11) (n=15) (n=11) Mean P Mean P Mean P

(95%CI) (95%CI) (95%CI)

Inflammation related

Anti-inflammatory

NFKBIA Nuclear factor kappa light 8.42 8.20 8.20 -0.220 0.085 -0.236 0.062 -0.016 0.894

polypeptide (0.09) (0.08) (0.09) (-0.47;0.03) (-0.48;0.01) (-0.26;0.22)

gene enhancer in B-cells

inhibitor, alpha

NFKBIZ Nuclear factor kappa light 3.41 3.29 3.27 -0.125 0.165 -0.141 0.115 -0.016 0.854

polypeptide gene enhancer (0.06) (0.05) (0.06) (-0.30;0.05) (-0.32;0.04) (-0.19;0.15)

in B-cells inhibitor, zeta

200

TNFAIP3 Homo sapiens TNF, alpha- 5.04 5.03 5.04 -0.007 0.095 -0.006 0.186 0.002 0.686

induced protein3 (0.03) (0.03) (0.03) (-0.02;0.01) (-0.01;0.03) (-

0.007;0.01)

FFAR3 Free fatty acid receptor 3 -2.18 -2.16 -2.18 -0.051 0.621 -0.071 0.483 -0.020 0.834

GPR41 (FFAR3) (0.07) (0.06) (0.07) (-0.26;0.16) (-0.27;0.13) (-0.22;0.18)

KRT1 Keratin 1, type II -0.715 0.047 -0.186 0.761 0.394 0.528 0.547 -0.233 0.783

(0.64) (0.54) (0.62) (-1.0;2.5) (-1.2;2.3) (-1.9;1.5)

Pro-inflammatory

IL6 Interleukin 6 (interferon, β -0.78 -0.61 -0.63 0.175 0.134 0.153 0.183 -0.022 0.838

2) (0.08) (0.07) (0.08) (0.06;0.41) (-0.08;0.38) (-0.24;0.20)

TNF Tumor necrosis factor 4.59 4.08 3.98 -0.510 0.048 -0.610 0.018 -0.100 0.677

(0.18) (0.15) (0.17) (-1.0;-0.005) (-1.1;-0.11) (-0.58;0.38)

IL1B Interleukin 1, beta 5.93 5.58 5.52 -0.350 0.092 -0.414 0.045 -0.065 0.739

(0.15) (0.12) (0.14) (-0.76;0.06) (-0.82;-0.01) (-0.46;0.33)

ICAM (log) Intercellular adhesion 0.37 0.35 0.35 -0.016 0.160 -0.021 0.075 -0.004 0.681

molecule 1 (0.08) (0.07) (0.08) (-0.04;0.007) (-0.04;0.002) (-0.03;0.02)

201

CCL3 Chemokine (C-C motif) 5.98 4.89 4.73 -1.086 0.121 -1.252 0.071 -0.167 0.799

ligand 3 (0.49) (0.42) (0.48) (-2.5;0.30) (-2.6;0.12) (-1.5;1.1)

CCL20 Chemokine (C-C motif) -1.42 -1.22 -1.29 0.193 0.016 0.128 0.098 -0.065 0.377

ligand 20 (0.05) (0.05) (0.05) (0.04;0.35) (-2.5;1.2) (-0.21;0.08)

CCL3L3 Chemokine (C-C motif) 2.29 2.13 2.00 -0.160 0.261 -0.288 0.044 -0.129 0.340

ligand 3-like 3 (0.10) (0.09) (0.10) (-0.44;0.12) (-057;-0.008) (-0.40;0.14)

CCL4L2 Chemokine (C-C motif) 5.81 5.94 5.94 0.129 0.104 0.130 0.097 0.001 0.990

ligand 4-like 2 (0.06) (0.05) (0.05) (-0.03;0.29) (-0.02;0.28) (-0.15;0.15)

CXCL1 Chemokine (C-X-C motif) -0.03 -0.29 -0.37 -0.286 0.150 -0.364 0.066 -0.078 0.674

ligand 1 (melanoma growth (0.14) (0.12) (0.13) (-0.68;0.11) (-0.75;0.02) (-0.45;0.30)

stimulating activityα)

CXCL2 Chemokine (C-X-C motif) 3.08 2.60 2.32 -0.485 0.180 -0.755 0.038 -0.270 0.428

ligand 2 (0.26) (0.22) (0.25) (-1.2;0.23) (-1.5;-0.04) (-0.95;0.41)

202

CXCL3 Chemokine (C-X-C motif) -0.39 -0.66 -0.71 -0.278 0.149 -0.324 0.091 -0.046 0.800

ligand 3 [ (0.14) (0.12) (0.13) (-0.66;0.19) (-0.70;0.05) (-0.41;0.32)

CXCL8 Chemokine (C-X-C motif) 1.43 0.859 0.220 -0.543 0.305 -1.182 0.028 -0.639 0.207

ligand 8 (0.30) (0.32 (0.36) (-1.6;0.52) (-2.2;-0.14) (-1.6;0.37)

EREG Epiregulin -0.10 -0.28 -0.30 -0.176 0.277 -0.192 0.229 -0.016 0.914

(0.11) (0.10) (0.11) (-0.50;0.15) (-0.51;0.13) (-0.32;0.29)

PTGS2 Prostaglandin-endoperoxide 1.05 0.57 0.13 -0.484 0.167 -0.922 0.010 -0.438 0.189

(COX2) synthase 2 (cyclooxygenase (0.25) (0.21) (0.24) (-1.2;0.21) (-1.6;-0.23) (-1.10;0.23)

RasGEF1B Ras GEF domain family 0.238 0.303 0.274 0.065 0.108 0.036 0.356 -0.029 0.450

member 1 B (0.03) (0.02) (0.03) (-0.01;0.14) (-0.04;0.011) (-0.10;0.05)

Cardiovascular disease related

FMN1 Formin 1 -3.23 -3.35 -3.42 -0.069 0.164 -0.063 0.199 0.006 0.893

(0.03) (0.03) (0.03) (-0.17;0.03) (-0.16;0.03) (-0.09;0.10)

203

Acute coronary syndrome

SELENBP1 Selenium binding protein 1 -2.19 -1.77 -2.51 0.352 0.338 -0.392 0.280 -0.744 0.038

(0.26) (0.22) (0.25) (-0.38;1.1) (-1.1;0.33) (-1.4;-0.04)

Obesity related

PPP1R15A Homo sapiens protein 2.93 2.76 2.72 -0.167 0.163 -0.210 0.077 -0.044 0.696

GADD34 phosphatase 1, regulatory (0.08) (0.07) (0.08) (-0.40;0.07) (-044;0.02) (-0.27;0.18)

subunit 15A

BCL2A1 BCL2-related protein A1 3.80 3.71 3.63 -0.087 0.421 -0.165 0.127 -0.078 0.449

(0.08) (0.06) (0.07) (-0.30;0.13) (-0.38;0.05) (-0.28;0.13)

Hemoglobin related

HBB Hemoglobin, beta 9.30 9.35 9.09 0.044 0.826 -0.213 0.289 -0.257 0.187

(0.14) (0.12) (0.14) (-0.36;0.45) (-0.61;0.19) (-0.64;0.13)

ALAS2 5'-aminolevulinate synthase 0.67 1.04 0.31 0.370 0.319 -0.357 0.330 -0.727 0.045

2 (0.27) (0.22) (0.26) (-0.37;1.1) (-1.1;0.38) (-1.4;-0.02)

HBA2 Hemoglobin, alpha 2 7.32 7.39 6.98 0.063 0.808 -0.341 0.185 -0.404 0.107

(0.18) (0.16) (0.18) (-0.46;0.58) (-0.85;0.17) (-0.90;0.09)

204

HBD Hemoglobin, delta 6.40 6.59 6.13 0.100 0.737 -0.360 0.223 -0.460 0.110

(0.21) (0.18) (0.20) (-0.50;0.60) (-0.95;0.23) (-1.0;0.11)

SLC4A1 Solute carrier family -3.69 -3.67 -4.07 0.019 0.958 -0.379 0.289 -0.398 0.250

4 (anion exchanger), (0.26) (0.22) (0.25) (-0.71;0.74) (-1.1;0.34) (-1.1;0.29)

member 1

Neuronal differentiation

FOXD4 Forkhead box D4 -3.33 -3.81 -3.84 0.028 0.361 -0.004 0.892 -0.032 0.273

(0.02) (0.02) (0.02) (-0.03;0.09) (-0.06;0.06) (-0.09;0.03)

CD83 CD83 molecule 4.65 4.50 4.50 -0.150 0.077 -0.145 0.083 0.005 0.945

(0.06) (0.05) (0.06) (-0.32;0.02) (-0.31;0.02) (-0.15;0.16

Cancer

Prostate

TMPO-AS1 TMPO antisense RNA 1 -2.27 -2.33 -2.33 -0.058 0.038 -0.053 0.053 0.005 0.854

long non-coding RNA (0.02) (0.01) (0.02) (-0.11;-0.03) (-0.11;0.01) (-0.05;0.06)

205

Esophageal squamous cell

CFLAR-AS1 CFLAR antisense RNA 1 -2.49 -2.51 -2.49 -0.019 0.277 -0.005 0.780 0.015 0.393

long non-coding RNA (0.01) (0.01) (0.01) (-0.05;0.02) (-0.04;0.03) (-0.02;0.05)

Gastric

LncRNA Long non-coding RNA G -3.38 -3.45 -3.43 -0.066 0.227 -0.046 0.392 0.020 0.696

GPR65-1 protein-coupled receptor 65- (0.04) (0.03) (0.04) (-0.18;0.04) (-0.15;0.06) (-0.08;0.12)

Unknown

PLA1A Phospholipase A1 member -3.03 -3.14 -3.14 -0.115 0.065 -0.111 0.071 0.004 0.942

A (0.04) (0.04) (0.04) (-0.24;0.007) (-0.23;0.01) (-0.11;0.12)

FAM53B Family with -2.55 -2.16 -2.17 0.087 0.011 0.074 0.025 -0.013 0.682

sequence similarity (0.02) (0.02) (0.02) (0.02;0.15) (0.01;0.14) (-0.07;0.05)

53, member B

SLED1 Proteoglycan 3 pseudogene 1.40 1.27 1.28 -0.129 0.090 -0.115 0.125 0.014 0.841

non-coding RNA (0.05) (0.04) (0.05) (-0.28;0.02) (-0.26;0.03) (-0.13;0.16)

GOS2 G0 /G1 switch gene 2 4.28 3.66 3.64 -0.620 0.171 -0.635 0.155 -0.015 0.972

206

(0.32) (0.27) (0.31) (-1.5;0.28) (-1.5;0.25) (-0.87;0.84)

BPIFB3 BPI fold containing family -2.46 -2.60 -2.58 - 0.132 0.032 -0.113 0.062 0.020 0.728

B, member 3 (0.04) (0.04) (0.04) (-0.25; -0.01) (-0.023;0.06) (-0.09;0.13)

BTBD19 BTB (POZ) domain -3.93 -3.83 -4.09 0.100 0.331 -0.157 0.129 -0.257 0.012

containing 19 (0.07) (0.06) (0.07) (-0.11;0.31) (-0.36;0.05) (-0.45;-

0.06)

LINC Long intergenic non-protein -3.47 -3.14 -3.37 0.330 0.040 0.100 0.516 -0.230 0.128

01220 coding RNA 1220 (0.11) (0.09) (0.11) (0.02;0.64) (-0.21;0.41) (-0.53;0.07)

KMT2E-AS1 KMT2E antisense RNA 0.90 0.74 0.43 -0.151 0.508 -0.465 0.044 -0.314 0.155

1(ncRNA) (0.16) (0.14) (0.16) (-0.61;0.31) (-0.92;-0.01) (-0.75;0.12)

ATP2B1-AS1 ATP2B1 antisense RNA 1 0.14 -0.04 -0.17 -0.177 0.218 -0.309 0.034 -0.131 0.338

(ncRNA) (0.10) (0.09) (0.10) (-0.46;0.11) (-0.59;-0.02) (-0.41;0.14)

LncRNA Long non-coding BTB -1.90 -2.04 -2.08 -0.142 0.145 -0.177 0.067 -0.036 0.696

BTBD19-1 domain containing 19_1 (0.07) (0.06) (0.07) (-0.33;0.05) (-0.37;0.01) (-0.11;0.31)

LncRNA Long non-coding Acyl-CoA -1.49 -1.86 -1.76 - 0.373 0.020 -0.272 0.080 0.101 0.493

ACOT-13 thioesterase 13 (0.11) (0.10) (0.11) (-0.68;-0.06) (-0.58;0.03) (-0.22;0.15)

207

LncRNA Long non-coding arresting -0.83 -0.87 -0.88 -0.039 0.718 -0.054 0.610 -0.015 0.882

ARRCD3-1 containing domain 3-1 (0.08) (0.07) (0.07) (-0.26;0.18) (-0.27;0.16) (-0.22;0.19)

LncRNA Long non-coding -2.30 -2.46 -2.19 -0.161 0.070 0.106 0.220 0.268 0.003

IYD iodotyrosine deiodinase (0.06) (0.05) (0.06) (-0.34;0.01) (-0.07;0.28) (0.10;0.43)

LncRNA Long non-coding lysine -1.85 -2.07 -2.16 -0.216 0.212 -0.307 0.076 -0.091 0.579

KDM5A-3 demethylase 5A-3 (0.12) (0.11) (0.12) (-0.56;0.13) (-0.65;0.03) (-0.42;0.24)

LncRNA Long non-coding small -0.52 -0.30 -0.37 0.220 0.004 0.146 0.047 0.074 0.285

SNRPD3-2 nuclear ribonucleoprotein (0.05) (0.04) (0.05) (0.07;0.37) (0.002;0.29) (-0.10;0.06)

D3 polypeptide-2

LncRNA Long non-coding solute 0.06 -0.27 -0.28 0.330 0.060 0.345 0.046 -0.016 0.923

SLC39A8-1 carrier family 39 member 8 (0.12) (0.11) (0.12) (-0.67;0.01) (-0.68;-0.007) (-0.34;0.31)

LncRNA Long non-coding Interleukin -1.28 -1.01 -1.09 0.284 0.066 0.189 0.224 -0.085 0.567

IL1B1 1, beta (0.11) (0.10) (0.11) (-0.04;0.89) (-0.12;0.50) (-0.39;0.21)

LncRNA Long non-coding NFKB -3.56 -3.49 -3.52 0.074 0.006 0.041 0.107 -0.033 0.178

NFKBID-1 inhibitor delta 1 (0.02) (0.02) (0.02) (0.02;0.12) (-0.009;0.09) (-0.08;0.02)

208

LncRNA Long non-coding -3.10 -2.93 -2.97 0.175 0.039 0.129 0.118 -0.046 0.557

PDE3B-1 phosphodiesterase 3B-1 (0.06) (0.05) (0.06) (0.009;0.34) (-0.03;0.29) (-0.20;0.11)

LncRNA Long non-coding marker of -2.52 -2.53 -2.54 -0.010 0.567 -0.021 0.232 -0.011 0.519

MKI67-2 proliferation Ki-67-2 (0.01) (0.01) (0.01) (-0.04;0.02) (-0.05;0.01) (-0.04;0.02)

LncRNA Long non-coding -2.76 -3.07 -3.07 -0.314 0.020 -0.310 0.020 0.003 0.979

CCT8-1 chaperoning containing (0.09) (0.08) (0.09) (-0.057;-0.05) (-0.057;-0.05) (-0.25;0.25)

TCP1 subunit 8

LncRNA Long non-coding leucine -1.91 -2.00 -1.98 -0.091 0.062 -0.073 0.125 0.018 0.694

LRRC61-2 rich repeat containing 61-2 (0.03) (0.03) (0.03) (-0.19;0.005) (-0.17;0.02) (-0.07;0.11)

LncRNA Long non-coding G protein- -3.00 -3.31 -3.47 -0.313 0.186 -0.470 0.048 -0.157 0.483

GRK3-1 coupled receptor kinase 3-1 (0.17) (0.14) (0.16) (-0.78;0.16) (-0.093;-0.005) (-0.61;0.29)

LncRNA Long non-coding RNA -2.37 -2.39 -2.38 -0.014 0.095 -0.011 0.179 0.003 0.706

TOMM20-2 translocase of outer (0.01) (0.01) (0.01) (-0.03;0.003) (-0.03;0.005) (-0.01;0.02)

mitochondrial membrane 20-

209

LncRNA RP11- Long non-coding RNA -2.35 -2.78 -2.82 -0.426 0.075 -0.466 0.050 -0.039 0.860

701P16.2.1-2 RP11-701P16.2.1-2 (0.17) (0.14) (0.16) (-0.90;0.04) (-0.93;-0.001) (-0.49;0.41)

LncRNA Long non-coding coiled-coil -2.00 -2.09 -2.12 -0.088 0.090 -0.118 0.024 -0.029 0.547

CCDC117-1 domain containing 117-1 (0.04) (0.03) (0.03) (-0.19;0.01) (-0.22;-0.02) (-0.13;0.07)

LOC101929524 Uncharacterized -3.31 -3.20 -3.25 0.106 0.030 0.057 0.223 -0.049 0.280

LOC101929524 (0.03) (0.03) (0.03) (0.01;0.20) (-0.04;0.15) (-0.14;0.04)

LOC644090 Uncharacterized -2.47 -2.64 -2.66 -0.163 0.087 -0.187 0.048 -0.024 0.789

LOC644090 (0.07) (0.06) (0.06) (-0.35;0.02) (-0.04;0.002) (-0.20;0.16)

Data expressed as mean ± standard error (SE) or mean (95% confidence interval, CI). *logarithmic transformation of the variable. Inter treatment comparisons by ANCOVA Model adjusted by sex and age. Significant differences values were set for p<0.05. Abbreviations: OJ, orange juice; EOJ, hesperidin-enriched orange juice.

210

Figure 1. Venn diagram of the number of differentially (P < 0.05) and borderline (P > 0.10) expressed genes in PBMC of stage 1 and prehypertensive individuals: 6 h after the consumption of 500 mL (A) and after 12-week consumption of orange juice (OJ) and hesperidin- enriched orange juice (EOJ) (B) compared to control drink, and the differentially expressed genes after both consumptions (C). Yellow, postprandial OJ; pink, postprandial EOJ; brown, sustained OJ; red, sustained EOJ.

211

Panel A

Panel B

Figure 3. Biological networks based on the differentially expressed gens after sustained consumption of orange juice (panel A) and hesperidin-enriched orange juice (panel B). Interaction between the differently expressed genes and other important related gens. Down- and up-regulated genes are symbolized in red and green color, respectively.

212

Figure 2. Real-time PCR validation of the arrays results for a subset of 6 genes.

213

Supporting Information Table S1. Composition of the three test drinks. Dose per day in mg/500 mL Control drink Orange juice Hesperidin-enriched orange juice Hesperidin ND 392 670 Narirutin ND 64 77 Vitamin C 235 235.3 235 Glucose 14407 938.2 9382 Fructose 7357 10123 10123 Sucrose 13197 18158 18158 Composition determined by State of Florida, Department of Citrus. Abbreviations: ND, no detected.

214

Supporting Information Table S2. Nucleotide sequences of primers used for PCR amplification. Forward primer Reverse primer Gene Ref. or Acc. No. (5’ to 3’) (5’ to 3’) DHRS9 TGGACCACGCTCTAACAAGT CACACTGCCTTGGGATTAGC NM_001142270.2 DSP TACCCCTGCGACAAGAACAT TCTGGGTTACGAGGCTTCAG NM_004415.4 IER3 GGCTTCTCTTTCTGCTGCTC GAGGGCTCCGAAGTCAGATT NM_003897.4 NAMPT TGGAGGAAGGAAAAGGAGACC CACACAACACACACCCAGTC NM_005746.3 PTX3 GGTAAATGGTGAACTGGCGG TGACAAGACTCTGCTCCTCC NM_002852.4 RPLP0 ATGGCAGCATCTACAACCCT AGGACTCGTTTGTACCCGTT NM_001002.4 TNF CAGAGGACCAGCTAAGAGGG TCAGTGCTCATGGTGTCCTT NM_000594.4

215

Supporting Information Table S3. Top canonical pathways, diseases and functions regulated by sustained consumption (12 weeks) of orange juice and hesperidin-enriched orange juice in PBMCs of pre- and stage 1 hypertensive individuals compared to control drink. Treat- Top canonical P- Top diseases and P-value ment pathways value functions range EOJ Granulocyte adhesion 8.85-17 Inflammatory response: 6.75-04– and diapedesis: CCL3, CXCL1, CXCL2, 4.17-16 CCL20, CCL3L3, CXCL3, CXCL8, IL1B, CCL4L2, CXCL1, PTGS2, PTX3, TNF, CXCL2, CXCL3, CCL20, CCL3, CXCL8, IL1B, TNF. CCL3L3, CD83, NFKB1A, NAMPT, DSP. Agranulocyte adhesion 1.71-16 Cell–to-cell signaling 6.75-04– and diapedesis: CCL3, and interaction: CCL20, 3.57-13 CCL20, CCL3L3, CCL3, CCL3L3, CCL4L2, CXCL1, CXCL1, CXCL2, CXCL2, CXCL3, CXCL3, CXCL8, IL1B, CXCL8, IL1B, TNF. PTGS2, TNF, CD83, PTX3, NFKB1A, DSP, NAMPT. Role of IL-17A in 8.20-12 Organismal injury and 9.41-04– arthritis: CCL20, abnormalities: CCL20, 2.42-12 CXCL1, CXCL3, CCL3, CD83, CXCL1, CXCL8, NFKB1A, CXCL2, CXCL3, PTGS2. CXCL8, IL1B, NAMPT, NFKB1A, PTGS2, TNF. TREM1 signaling: 3.68-11 Hematological system: 8.86-04– CCL3, CD83, CXCL8, CCL20, CCL3, 3.12-12 IL1B, TNF. CCL3L3, CXCL1, CXCL2, CXCL3, CXCL8, IL1B, TNF, CD83, NFKB1A, PTX3. Communication 1.40-10 Cardiovascular system: 7.36-04– between innate and CCL3, CXCL1, 2.53-09 adaptive immune cells: CXCL2, CXCL8, IL1B, CCL3, CCL3L3, CD83, PTGS2, PTX3, TNF, CXCL8, IL1B, TNF. NFKB1A.

OJ Dendritic cell 1.84-07 Inflammatory response: 1.39-03– maturation: CD83, CCL20, CD83, IL1B, 7.24-11 IL1B, NFKB1A, TNF. NFKB1A, TNF. IL-23 signaling 3.61-07 Organismal injury and 1.39-03– pathway: IL1B, abnormalities: CCL20, 7.24-11 NFKB1A, TNF.

216

CD83, IL1B, NFKB1A, TNF. IL-10 signaling: IL1B, 1.33-06 Cellular growth and 1.22-03– NFKB1A, TNF. proliferation: CCL20, 3.21-08 CD83, IL1B, NFKB1A, TNF. TREM1 signaling: 1.39-06 Lipid metabolism: 1.04-03– CD83, IL1B, TNF. CD83, IL1B, NFKB1A, 3.98-08 TNF, PLA1A. Toll-like receptor 1.71-06 Hematological system: 1.39-03– signaling: IL1B, CD83, IL1B, NFKB1A, 3.00-08 NFKB1A, TNF. TNF, CCL20.

Analysis performed with the Ingenuity Pathway Analysis (IPA). Abbreviations: PBMCs, peripheral blood mononuclear cells; EOJ, hesperidin-enriched orange juice; OJ, orange juice.

217

Supporting information Table S4. Statistically significant and borderline differentially expressed genes after the single dose study and after 12 weeks of OJ, EOJ and CD consumption. After 6h After 12 weeks (single dose (sustained study) consumption) Genes OJ EOJ EOJ OJ EOJ EOJ vs vs vs vs vs vs CD CD OJ CD CD OJ Anti-inflammatory

NFKB1A ↓b ↓ b ↓ b ↓ b

TNFAIP3 ↓a ↓ b Pro-inflammatory TNF ↓a ↓a IL1B ↓ b ↓a

ICAM ↓ b CCL3 ↓b ↓ b CCL20 ↑a ↑ b

CCL3L3 ↓ b ↓a CCL4L2 ↓ b ↓ b ↑ b

CXCL1 ↓ b CXCL2 ↓ b ↓a CXCL3 ↓ b CXCL8 ↓ a CLEC5 ↑ b

PTGS2 ↓a Insulin resistance DHRS9 ↑a ↓a

Acute coronary syndrome SELENBP1 ↓a

218

Endothelial function related DSP ↑a

IER3 ↓a

Improvement of hypertension PTX3 ↓a

NAMPT ↓b

Obesity related

PPP1R15AGADD34 ↓b

Hemoglobin related ALAS2 ↓a

Neuronal differentiation CD83 ↓b ↓b

Cancer: prostate TMPO-AS1 ↓a ↓b

Unknown function SLED1 ↓ b ↓b

BPIFB3 ↓a ↓b

PKDL1 ↑ a

ATP2B1-AS1 ↓a

PLA1A ↓b ↓b

FAM53B ↑a ↑a

BTBD19 ↓a

KMT2E-AS1 ↓a

LINC01220 ↑a

LncRNA ATOH8-2 ↑ b

LncRNA ACOT-13 ↓a ↓b

LncRNA KDM5A-3 ↓b

LncRNA SNRPD3-2 ↑a ↑a

219

LncRNA SLC39A8-1 ↑b ↑a

LncRNA CCT8-1 ↓a ↓a

LncRNA GRK3-1 ↓a

LncRNA RP11- ↓b ↓b 701P16.2.1-2 LncRNA CCDCC117-1 ↓b ↓a

LOC101929524 ↑a

LOC644090 ↓b ↓a

LncRNA IYD ↓b ↑a

LncRNA BTBD19-1 ↓b

LncRNA IL1B1 ↑b

LncRNA NFKBID-1 ↑a

LncRNA PDE3B-1 ↑a

LncRNA LRRC61-2 ↓b

LncRNA TOMM20-2 ↓b a, p<0.05; b, p<0.10. Abbreviations: OJ, orange juice; EOJ, hesperidin- enriched orange juice; CD, control drink.

220

Project 4: The consumption of hesperidin in orange juice modulates the plasma and urine metabolic profiles in pre- and stage 1 hypertensive subjects to promote beneficial effects on the cardiovascular system: targeted and nontargeted metabolomic approaches (CITRUS study) (Editor submitted)

221

222

The consumption of hesperidin in orange juice modulates the plasma and urine metabolomic profiles in pre- and stage 1 hypertensive subjects to promote beneficial effects on the cardiovascular system: targeted and nontargeted metabolomic approaches (CITRUS study)

Pla-Pagà L1,2, Pedret A1,2,a*, Valls RM2,1,a*., Calderón-Pérez L1,2, Llauradó E2, Companys J1,2, Martín-Luján F2,3,4, Moragas A2,3,5, Canela N1,6, Puiggròs F1, Caimari A1, del Bas JM1, Arola L1,a, Solà R1,2,7,a, Mayneris-Perxachs J1,8,a.

aSenior author 1Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain. 2Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut, Functional Nutrition, Oxidation and Cardiovascular Disease Group (NFOC-Salut), Reus, Spain. 3Institut Universitari d’Investigació en Atenció Primària-IDIAP Jordi Gol, Tarragona, Spain. 4Primary Care Centre Sant Pere, Institut Català de la Salut, Tarragona, Spain. 5Primary Care Centre Jaume I, Institut Català de la Salut, Tarragona, Spain. 6Eurecat, Centre Tecnològic de Catalunya, Centre for Omic Science, Reus, Spain. 7Hospital Universitari Sant Joan, Reus, Spain.

223

8Department of Endocrinology, Diabetes and Nutrition, Dr Josep Trueta University Hospital, Girona Biomedical Research Institute (IDIBGI), Girona, Spain.

Keywords: metabolomic analysis, plasma, urine, hesperidin, orange juice, hypertensive.

*Corresponding authors: Anna Pedret Figuerola, PhD. Eurecat, Centre Tecnològic de Catalunya. Av/ Universitat 1, 43204 Reus, Spain. E-mail: [email protected] Rosa M Valls Zamora, PhD. Universitat Rovira i Virgili, Facultat de Medicina i Ciències de la Salut. C/ Sant Llorenç 21, CP 43201 Reus, Spain. Email: [email protected]

Abbreviations CD: control drink CVDs: cardiovascular diseases CVDRFs: cardiovascular diseases risk factors DMA: dimethylamine DMG: dimethylglycine EOJ: hesperidin-enriched orange juice GPC: glycerophosphocholine NAG: N-acetylglycoproteins OJ: orange juice PAG: phenylacetylglutamine 2-HIB: 2-hydroxyisobutyrate

224

ABSTRACT Scope The aim of the present work was to determine new biomarkers of the biological effects of hesperidin in orange juice (OJ) applying a nontargeted metabolomics approach validated by targeted metabolomics analyses of compliance biomarkers. Methods and results Plasma/serum and urine targeted (HPLC-MS/MS) and untargeted (1H- NMR) metabolomics signatures were explored in a subsample with pre- and stage-1 hypertension subjects of the CITRUS study (N=159). Volunteers received 500 mL/day of control drink, OJ, or hesperidin- enriched OJ (EOJ) for 12-weeks. A 6-h postprandial study was performed at baseline. Targeted analyses revealed plasma and urine hesperetin 7-O- -D-glucuronide as the only metabolite differing between OJ and EOJ groups after both single dose and 12-weeks consumption, and in urine is correlated with a decreased systolic blood pressure level. The nontargeted approach showed that after single dose and12-weeks consumption of OJ and EOJ changed several metabolites related with an anti-inflammatory and antioxidant actions, lower blood pressure levels and uraemic toxins. Conclusions Hesperetin 7-O- -D-glucuronide could be a candidate marker for distinguishing between the consumption of different hesperidin doses as well as a potential agent mediating blood pressure reduction. Moreover, changes in different endogenous metabolites could explain the mechanisms of action and the biological effects of hesperidin consumption.

225

1. INTRODUCTION The metabolome comprises all the metabolites found in an organ, cell, biofluid or organism under certain conditions. Metabolites have several functions in cells and systems,(69) and their levels in organism are directly associated with human health.(163) Notably, changes at the metabolome level can appear in biofluids before the appearance of clinical symptoms.(164) Therefore, when assessing nutritional studies, metabolomics approaches provide information about changes in diet- derived and endogenous metabolites to allow the identification of novel biomarkers related to dietary intake and biological effects.(165)

The relationship between bioactive compounds in food and the development of several chronic conditions has been widely studied. In this sense, the effects of phenolic compounds, that are naturally present in plant-based foods, have emerged as potential nutritional strategy against on cardiovascular disease (CVD). In particular, hesperidin, which is the main flavanone in citrus fruits and is present mainly in oranges and orange juice (OJ), has demonstrated beneficial effects on different cardiovascular risk biomarkers in animal and human studies.(130) Nevertheless, the bioavailability of the flavanone hesperidin depends on the body’s capacity to absorb this metabolite, and subjects can be stratified into three categories: low, moderate and high absorbers.(110) Through omics analyses, our research group recently showed that hesperidin consumption can modify the heart and kidney tissue proteome(166) and the plasma metabolome profile of rats with metabolic syndrome.(167) Specifically, our metabolomics analysis showed that hesperidin consumption decrease the levels of several

226 plasma metabolites related to inflammation and oxidative stress and thereby, reduces the blood pressure levels, influences on the cardiovascular system, and improves the conditions of metabolic syndrome in rats.(167) In humans, hesperidin consumption decreases the body weight(94) and diastolic blood pressure levels,(107) and improves lipid profile(93) and postprandial microvascular endothelial reactivity.(107) In fact, our research group recently demonstrated a beneficial effect of hesperidin consumption in OJ on blood pressure levels in subjects with pre- and stage 1 hypertension.(133)

The study of the metabolic profiles can offer important insights into cardiovascular system and CVDs pathogenesis as well as the possible identification of new CVDs biomarkers. However, to the best of our knowledge, no study has combined a targeted and nontargeted metabolomics approach after the consumption of hesperidin in OJ in humans with cardiovascular disease risk factors (CVDRFs). Therefore, the aim of the present work was to determine new biomarkers of the biological effects of hesperidin in OJ applying nontargeted metabolomics approach in plasma/serum and urine samples after both single dose (6 hours) and sustained (12 weeks) consumption, validated through targeted metabolomics analyses of compliance biomarkers in subjects with pre- and stage 1 hypertension.

2. METHODS 2.1 Study population and study design The study population was obtained from the CITRUS study (N= 159), a randomized, parallel, double-blind and placebo controlled clinical trial (registration number in Clinical-Trials.gov: NCT02479568).(133)

227

Moreover, the targeted metabolomics analysis was performed in 129 subjects who completed sustained study, and nontargeted metabolomics was performed in 52 subjects who completed both the single dose and sustained studies. The subjects were women and men with pre- or stage 1 hypertension, systolic blood pressure ≥ 120 mm Hg and < 159 mm Hg and diastolic blood pressure < 100 mm Hg, who are not taking any anti-hypertensive or anti-hyperlipemia medication. The subjects, who were recruited between January 2016 and June 2017, were randomly assigned to one group of the three intervention groups: control drink (CD), OJ and hesperidin-enriched orange juice (EOJ) groups. Detailed information on the study population and the design of the study are described in Supplementary Information Methods S1.

2.2 Test drinks The following three test drinks were provided by The Florida Department of Citrus from the USA: the CD (without hesperidin), OJ containing hesperidin at its natural level, and EOJ, which was OJ enriched with hesperidin. The intervention drinks were similar in smell and appearance and can be differentiated by a code assigned by an independent researcher not related directly to the clinical trial.

2.3. Dosage information All subjects drank 500 mL of the corresponding test drink daily: the CD, provided 0 mg/day of hesperidin, OJ provided 345 mg/day of hesperidin, and EOJ provided 600 mg/day of hesperidin. The composition of the three test drinks is detailed in Supporting Information Table S1.

228

2.4 Targeted metabolomics analysis of compliance biomarkers Sample collection To assess whether the compliance biomarkers in plasma supply more information than those in urine, plasma samples were collected at baseline and after 2, 4 and 6 h in the single dose study. In the sustained study, plasma samples were collected at baseline and at 4, 8 and 12 weeks after sustained consumption. Moreover, 24-h urine samples were collected at the beginning of the study and after 12 weeks of treatment.

Determination of hesperidin and naringenin metabolites in urine and plasma samples by UHPLC-MS/MS Urine and plasma samples were thawed at 4 °C. Subsequently, 50 µL of the urine samples was mixed with 100 μL of water containing 1% HF or the internal standard (rac-hesperetin-d3). This dilution was performed through a semi-automated process using the Agilent Bravo Automated Liquid Handling Platform. In contrast, 20 internal standards (hesperetin-d4) were mixed with 125 μL of the plasma samples and 750 μL of methanol, and then, mixture was then vortexed for 10 mins and centrifuged at 4700 rpm at 4°C for 10 min. A total of 900 μL was transferred and evaporated in a Speed-Vac at ambient temperature. The residues were reconstituted in 25 μL of MeOH and 75 μL of H2O (1% HFor) and injected into an LC-MS/MS instrument. The extraction was performed through a semi-automated process using the Agilent Bravo Automated Liquid Handling Platform. The analyses were performed with an Agilent 1200 series UHPLC coupled to a 6490 Triple Quad LC- MS mass spectrometer, and the source electrospray source ionization (ESI) was operated in the negative mode. A reversed-phase Eclipse Plus column (C18, 1.8 μm, 2.1 mm × 150 mm) from Agilent Technologies was used for chromatographic separation.

229

The assignment of the hesperetin and naringenin metabolites was performed by direct comparison with commercial standards (hesperidin, narirutin, hesperetin, naringenin, hesperetin 3-O-β-D- glucuronide, hesperetin 7-O- β -D-glucuronide, naringenin 4'-O-β-D- glucuronide and hesperetin 7-O-sulfate). The tentative identification of the other hesperetin and naringenin metabolites was based on the precursor ion mass, chromatographic behaviour on a similar system, and published data from plasma samples. The method was validated using a pool of samples collected during the study and was based on standard addition. The method was validated by determining the limit of detection (MDL), and the limit of quantification (MQL), the repeatability (expressed as relative standard deviation, RSD), the accuracy (%) and the coefficient of determination of the calibration curve (Supplementary Information Table S2). The total flavanone content was obtained by summing the plasma or urine levels of all identified hesperidin and naringenin metabolites.

2.5 Nontargeted metabolomics analysis of biological effects biomarkers Sample collection Serum samples were collected at baseline and 6 h after treatment in the single dose study (N=129). Nontargeted metabolomics analyses were performed at 6-h, corresponding with the reported maximal flavanone plasma concentration.(168) For the sustained study, serum samples were collected at the beginning and after 12 weeks of treatments only for those subjects that participated in single dose study (N=52). Twenty- four hours urine samples were also collected at the beginning of the study and after 12 weeks of treatments (N=129).

230

NMR sample preparation and acquisition Urine and serum samples were allowed to reach room temperature. Subsequently, 400 μL of urine was mixed with 200 μL of phosphate buffer (pH 7.4; 100% D2O) containing the internal standard/L, 3- trimethylsilyl-1-[2,2,3,3-2H4] propionate (TSP), at a concentration of 1 mM and 2 mM sodium azide. In addition, 200 μL of each serum sample was mixed with 400 μL of phosphate buffer containing 2 mM TSP, and the resulting mixture was vortexed and centrifuged at 10,000 x g for 10 min. A 550-μL aliquot was then transferred into a 5-mm NMR tube. All NMR spectra were recorded at 300 K with a Bruker Avance III 600 spectrometer operating at 600.20 MHz for protons and equipped with a 5-mm PABBO BB-1H/D Z-GRD probe and an automatic sample changer with a cooling rack at 4°C. For the urine samples, a standard one-dimensional (1D) NOESY presaturation pulse sequence (RD-90°- t1-90°-tm-90°-acquire) was used with water suppression. The data from all the samples were acquired using a recycle delay (RD) of 5 s, a mixing time (tm) of 100 ms, an acquisition time of 3.41 s, and a 90° pulse of 10.99 μs. For each sample, 128 scans were collected after four dummy scans to obtain 64K data points with a spectral width of 16 ppm. For all serum samples, 1D-NOESY and Carr-Purcell-Meiboom-Gill (CPMG) spin echo experiments with water suppression were performed to obtain the corresponding metabolic profile. The settings for the NOESY experiments were the same as those used for the urine samples. Low-molecular-weight-filter CPMG spectra were then obtained using a CPMG sequence (RD [90°x-(t-180°y-t)n-collect FID) with a spin- echo delay of 400 μs (for a total T2 filter of 210 ms), which allowed efficient attenuation of the lipid NMR signals. The total acquisition time

231 was 2.73 s with an RD of 2 s, and the 90° pulse length was automatically calibrated for each sample at approximately 11.1 μs. For each sample, eight dummy scans were followed by 128 scans, and 64K points over a spectral width of 20 ppm were collected.

NMR spectra processing Prior to the data analysis, all FIDs were multiplied by an exponential function equivalent to a 0.3-Hz line broadening factor before applying a Fourier transform. The transformed spectra were phased, corrected for baseline distortions, and calibrated using the reference standard TSP in Topspin 3.5 (Bruker). Each spectrum in the range between 0 and 10.0 ppm was digitized into consecutive integrated spectral regions (32.768) of equal width (0.00033 ppm) using MATLAB (MathWorks). The region containing the residual water in both the urine and serum samples was removed to minimize the baseline effects arising from improper water suppression. Chemical shift variation was minimized across the dataset by applying a recursive segment-wise peak alignment algorithm to each spectrum. Each spectrum was then normalized using a probabilistic quotient normalization.

2.6 Statistical analysis To express the changes in plasma compliance biomarkers after 4, 8 and 12 weeks of the treatments, intra-treatment comparisons were evaluated by a general linear model with Bonferroni correction and age and sex as covariables, and inter-treatment comparisons were evaluated with the ANCOVA model adjusted by sex and age. To express the changes in urine compliance biomarkers after 12 weeks of the treatments, intra- treatment comparisons were evaluated by Wilcoxon test and inter- treatment comparison by Mann-Whitney test for independent samples.

232

The relationship of hesperetin 7-O-B-glucuronide and systolic blood pressure levels was assessed by Pearson's and Spearman's correlation coefficients since this metabolite seems to be the responsible of hypotensive, vasodilatory and anti-inflammatory activities in literature.(15)

Multivariate modelling of the results from the nontargeted metabolomics analysis was performed in MATLAB using in-house scripts. First, outlying samples were identified by PCA, and the significant metabolites that can be used to discriminate between groups were identified using orthogonal partial least squares discriminant analysis (O-PLS-DA). Hence, NMR metabolic profiles served as the descriptor matrix (X) and the experimental groups were used as the response variable (Y). The O-PLS model loadings were back- transformed by multiplying all the values by their standard deviation (covariance) and colour-plotted based on their model weights (R2). The important variables for between-class discrimination are highlighted by the correlation colour scale, and the red colour indicates high significance. The predictive performance (Q2Y) of the models was calculated using a 7-fold cross-validation approach, and the model significance was assessed using 1000 permutations. Large interindividual variability has been observed in the bioavailability and excretion of hesperidin among subjects.(169,170) Therefore, we also build O-PLS models using the maximum total flavanone (hesperidin + naringenin metabolites) content, independently from the intervention group, as a continuous predictor variable (Y).

233

3. RESULTS 3.1 Volunteer characteristics and results from the human randomized clinical trial Of the 159 volunteers with pre- and stage 1 hypertension who were randomly allocated, 129 completed the sustained study (N = 43, 46 and 40 in CD, OJ and EOJ groups, respectively), and 52 of these volunteers participated in the baseline single-dose study (N = 17, 21 and 14 in the CD, OJ and EOJ groups, respectively). The flow diagram of the volunteers is detailed in Figure S1 in the Supplementary Information, and the basal characteristics of the volunteers are detailed in Table S3 in the Supplementary Information.

3.2 Targeted metabolomics for compliance biomarkers Plasma targeted metabolomics analysis The changes in the six studied plasma metabolites, namely, hesperetin 7-O-β-D-glucuronide, hesperetin 3-O-β-D-glucuronide, hesperetin 7- O-sulfate, naringenin 4-O-β-D-glucuronide, naringenin glucuronide and naringenin sulfate, after 4, 8 and 12 weeks of the treatments are detailed in Table 1 (N = 129). Compared with those observed in the CD group, significantly increased levels of the six exposure biomarkers were observed after 12 weeks of OJ and EOJ consumption. In particular, the metabolite hesperetin 7-O-B-D-glucuronide was the main metabolite that showed differential expression between the OJ and EOJ groups. The changes in the six studied plasma metabolites, namely, hesperetin 7-O-B-D-glucuronide, hesperetin 3-O-B-D-glucuronide, hesperetin 7- O-sulfate, naringenin 4-O-B-D-glucuronide, naringenin glucuronide and naringenin sulfate, detected after 2, 4 and 6 h in the single dose study are presented in Figure 1 (N = 52). After 6 h, OJ and EOJ

234 consumption significantly increased the plasma levels of hesperetin 7- O-B-D-glucuronide, hesperetin 3-O-B-D-glucuronide, hesperetin 7-O- sulfate, naringenin glucuronide and naringenin sulfate compared with the levels obtained with the CD. High interindividual variability has been observed in absorption and excretion in hesperidin-based acute and chronic interventions.(112) Consistently, we also observed a large interindividual variability in total plasma flavanone pharmacokinetics upon beverage intake in each treatment group (Supplementary Information Figure S2). Some participants had the maximal total flavanone concentration after 6-h of consumption, while other had the maximum peak at 4-h or even earlier (2-h). In addition, the total amount of flavanones absorbed in each hesperidin supplemented group had a high variability, with high-, medium-, and low-absorbers in each group.

Urine targeted metabolomics analysis: The changes in the six studied urine metabolites, namely, hesperetin 7- O-β-D-glucuronide, hesperetin 3-O-β-D-glucuronide, hesperetin 7-O- sulfate, naringenin 4-O-β-D-glucuronide, naringenin glucuronide and naringenin sulfate, after sustained consumption of the intervention drinks are detailed in Table 2 (N = 129). After 12 weeks of CD intervention, no differences were observed in the six studied urine metabolites. Compared to CD, after OJ and EOJ interventions, significant increased levels of the six studied metabolites were observed. Moreover, the EOJ consumption compared to OJ, increased the levels of only one metabolite in urine: the hesperetin 7-O-β-D- glucuronide. In Figure 2 is represented the relationship between the levels of urine hesperetin 7-O-β-D-glucuronide and changes on blood pressure levels

235 of the participants of CITRUS study. Our results showed that increased levels of hesperetin 7-O-β-D-glucuronide in urine are correlated with lower blood pressure levels after 12 weeks of treatment (R=-0.223; P=0.011).

3.3 Nontargeted metabolomics analysis of biomarkers of the biological effect in the serum and urine samples Pairwise O-PLS-DA models were built to compare the metabolic profiles of the three study groups at 6-h in the single dose study. Significant models were obtained for the comparison of the metabolic profiles between CD and OJ group (Q2Y=0.36, P=0.001) and between CD and EOJ groups (Q2Y=0.35, P=0.001) after the 6 hours of the single dose consumption. The serum levels of proline betaine and dimethylglycine (DMG) were significantly increased and the level of leucine were significantly decreased after 6 h, in the OJ and EOJ groups compared with the CD group. Moreover, 6 h after EOJ consumption, the isoleucine urine levels were also significantly decreased compared with those in the CD group (Figure 3, panels A and B). To take into account that hesperidin absorption varied largely in both OJ and EOJ groups (Figure 3, panel C), we also built an O-PLS model considering the the maximal total flavanone concentration observed during the single dose study for each participant independent of the treatment group. Interestingly, we obtained a more significant model (Q2Y=0.20, P<0.001) that revealed additional differences between high and low absorbers (Figure 3, panel D). Consistent with the previous models, high hesperidin absorbers had higher plasma levels of DMG and proline betaine. However, this model also revealed that high absorbers had higher levels of ketone bodies (3-hydroxybutyrate, acetoacetate) but lower levels of BCAA, alanine, lysine, and glutamine.

236

Pairwise O-PLS-DA models were also constructed to compare the metabolic profiles between treatment groups at the end of the study (week 12) for those participants that completed the single dose study. Again, we obtained significant models differentiating the metabolic profiles of CD vs OJ (Q2Y=0.35, P=0.02) and CD vs EOJ (Q2Y=0.28, P=0.05) (Figure 4). After the sustained consumption study, a nontargeted metabolomics was performed with a subsample of 52 subjects (N = 17 in CD, 21 in OJ and 14 in EOJ) who completed both the single dose study and the sustained study, and differences in the metabolic profiles were observed after sustained consumption. As a result, increased serum levels of proline betaine and decreased serum levels of glycerophosphocholine (GPC), N-acetyl glycoproteins (NAGs), acetate, valine, isoleucine, and leucine were observed after 12 weeks of OJ consumption compared with the levels founded in the CD group (Figure 4, panel A). Similarly, increased serum levels of proline betaine and decreased plasma levels of GPC, aspartate, glutamate, valine, isoleucine, and leucine were observed after 12 weeks of EOJ consumption compared with the levels detected in the CD group (Figure 4, panel B).

The nontargeted analysis of urine samples from the sustained study (N=129) also revealed significant differences between the metabolic profiles of the OJ and CD (Q2Y=0.79, P<0.001) and between the EOJ and CD (Q2Y=0.80, P<0.001) groups after 12 weeks. Significantly increased levels of proline betaine in urine were observed after 12 weeks of OJ consumption compared with the levels found in the CD group, whereas the levels of 4-hydroxyhippurate, pseudouridine, phenylacetylglutamine (PAG), 4-cresyl sulfate, creatinine,

237 dimethylamine (DMA), NAGs, alanine and 3-methyl-2-oxovalerate were significantly decreased (Figure 5, panel A). Compared with the levels found in the CD group, significantly increased levels of proline betaine in urine were observed after 12 weeks of EOJ consumption, and the levels of 4-cresyl sulfate, pseudouridine, uracil, creatinine, creatine, NAGs, alanine, 2-hydroxyisobutyrate (2-HIB) and 3-MOV were significantly decreased (Figure 5, panel B). No differences were found between the OJ and EOJ groups in any of the untargeted analyses.

4. DISCUSSION The current study comprises the first evaluation of the effects on the metabolome of single dose and sustained consumption of hesperidin in OJ and EOJ on subjects with pre- and stage 1 hypertension through targeted and nontargeted approaches to finally determine new biomarkers of the biological effects of hesperidin. Both targeted and nontargeted analyses revealed that OJ and EOJ consumption induced significant changes on plasma/serum and urine metabolome compared to the CD group.

The targeted metabolomics approach for compliance biomarkers revealed increased plasma and urine of six flavanone catabolites after 12 weeks of OJ and EOJ consumption and 6 h after consumption of the single dose. As previously reported in plasma(133) and in the present work also in urine, the six compliance biomarkers were significantly increased after OJ and EOJ consumption, which demonstrate compliance with the interventions Moreover, hesperetin 7-O-B-D- glucuronide, both in plasma and urine, was the only metabolite that

238 showed differences between the OJ and EOJ groups, which indicated that hesperetin 7-O-B-D-glucuronide could be a candidate marker for distinguishing between different doses of hesperidin consumed for at least 12 weeks. Additionally, urine levels of hesperetin 7-O-B-D- glucuronide was correlated with lower systolic blood pressure levels, demonstrating a possible role on blood pressure control in humans. In this sense, its appreciation was only previously observed in rat models, where hesperetin 7-O-B-D-glucuronide exerts antihypertensive effects and exhibits vasodilatory and anti-inflammatory activities.(15) Thus, to the best of our knowledge, this is the first study to show a negative correlation between the urine metabolite hesperetin 7-O-B-D- glucuronide and the levels of systolic blood pressure.

Metabolic profiles provide a direct functional read-out of the physiological status of an individual. As a result, they are closely related to the phenotype. Therefore, we also applied an NMR-based untargeted metabolomics approach to identify metabolites and metabolic pathways associated with the phenotypes after hesperidin consumption to understand the mechanisms underlying these phenotypes. We found a significant increase in plasma and urine levels of proline betaine after both the single dose study and sustained consumption of OJ and EOJ compared to CD. Proline betaine is a direct marker of citrus fruit intake and it reflects a good treatment adherence.(171) Moreover, the levels of leucine, valine and isoleucine were significantly decreased in the serum and urine samples collected after 12 weeks of sustained consumption of OJ and EOJ and after consumption of a single dose of OJ and EOJ, which supported the effect of hesperidin consumption independent of the dose. Valine, leucine and isoleucine are branched-chain amino acids

239

(BCAA) associated with several cardiometabolic risk factors such as insulin resistance,(125) obesity, atherogenic dyslipidaemia, elevated blood pressure,(172,173) and increased incident of CVDs events.(124) The effects of BCAA in general, and leucine in particular, on mTOR activation could partly explain these associations.(172,173) Intracerebroventricular administration of leucine has shown to active the mTORC1 signalling pathway specifically in the mice hypothalamus and cause a significant increase in arterial pressure.(174) In this sense, previous studies have reported elevated circulating leucine levels in hypertensive subjects.(175) Notably, the circulating levels of BCAA, and leucine in particular, were significantly decreased after 12 weeks of OJ or EOJ consumption compared to the CD. In addition, the urinary levels of 3-methyl-2-oxovalerate (3-MOV), a microbial-derived metabolite from leucine degradation, were also decreased after chronic consumption of both OJ and EOJ treatments. These findings could partly explain the decrease in systolic blood pressure that we observed in the participants consuming OJ or EOJ. Therefore, decreased plasma levels of BCAA could point to a possible protective role of hesperidin in OJ on hypertension. Another potential mechanism by which hesperidin could reduce systolic blood pressure involves an improvement of the inflammatory and oxidative stress status. We observed decreased serum and urine levels of N-acetyl glycoproteins (NAGs) after 12 weeks of OJ and EOJ consumption. NAGs is considered a novel biomarker of systemic inflammation and cardiovascular disease risk.(176) In addition, the levels of pseudouridine, a metabolite used as a biomarker of oxidative stress in DNA,(177) were decreased in the urine samples collected after 12 weeks of OJ and EOJ consumption. Higher pseudouridine excretion

240 increases oxidative stress and inflammatory processes.(178) As a result, hesperidin in OJ and EOJ could improve the inflammatory state of subjects with pre- or stage 1 hypertension. Moreover, decreased plasma levels of GPC, which are related to decreases in oxidative and inflammatory tissue damage, were also observed after 12 weeks of OJ and EOJ consumption.(179) However, few studies have broadly discussed its role as a CVD risk factor. Therefore, decreased levels of NAG, pseudouridine and GPC, following hesperidin consumption reflects an improvement in the oxidative and inflammatory status of pre- and stage 1 hypertensive subjects, which could contribute to the amelioration of systolic blood pressure.

Hesperidin consumption also resulted in significant alterations in choline metabolism. Choline is rapidly absorbed in the small intestine and catabolized to betaine. Betaine plays a pivotal role in the one- carbon metabolism as a methyl-donor by reacting with homocysteine to generate methionine and DMG. However, gut bacteria compete with the host for choline. Hence, choline can be alternatively metabolized by the gut microbiota to TMA and dimethylamine (DMA), thereby depleting metabolites involved in one-carbon metabolism. Alterations in one-carbon metabolism has been related with metabolic diseases and hypertension-related health outcomes.(180) Notably, OJ and EOJ participants had higher plasma levels of DMG and lower urinary excretions of DMA, which suggests a lower microbial metabolism of choline following hesperidin consumption, that could explain the improvement of systolic blood pressure in these patients. DMA is also a microbial-derived uremic toxin that has shown to contribute to CVD(181) and is considered a potential uraemic

241 cardiovascular toxin according to the European Uraemic Toxin Working Group.(182) Other uraemic toxins, namely, 4-cresyl sulfate, creatinine and PAG, were significantly decreased in urine after OJ and EOJ consumption compared with their levels in the CD group. Lower serum levels of these toxins are beneficial to the cardiovascular and renal systems because the accumulation of their metabolites could produce vascular inflammation, endothelial dysfunction and vascular calcifications.(183) Moreover, PAG have also been negatively associated with systolic blood pressure (184) and with increased risk of incident of coronary artery disease.(185) In this sense, the lower urinary excretion of uraemic toxins observed after 12 weeks of OJ and EOJ consumption could be due to the minor serum levels derived from their lower production, which is likely related to an improvement in oxidative stress.(186) Thus, the lower oxidative stress obtained after OJ and EOJ consumption could result in the production of uraemic toxins.

Additionally, our results suggest that the individual capacity of flavanone absorption is an important aspect to consider. We observed a large interindividual variability in flavanone absorption, as reported in the literature,(111) and identify high and low total flavanone absorbers regardless of their consumption of OJ or EOJ. We considered the maximum absorption values of total flavanones in all the samples regardless of their intervention group at the end of the single dose study (6 h) and found that the subjects who absorbed more flavanones showed significant changes in their proline betaine, 3-hydroxybutyrate, leucine, isoleucine, valine, lysine, and alanine levels compared with the levels found in the CD group. This subanalysis was performed because a large interindividual variability was observed after the single-dose study,

242 which resulted in the identification of high and low absorbers in the OJ and EOJ groups. Moreover, this large interindividual variability in the absorption of flavanones, which includes the group of phenolic compounds to which hesperidin belongs, could at least partly explain why similar and non-significant changes in biomarkers of the biological effects of hesperidin were observed after OJ and EOJ consumption, although the doses of hesperidin consumed were different. Consistently, the volunteers who participated in the single dose study (N = 52) and presented higher plasma levels of total flavanones, showed higher plasma levels of ketone bodies, such as 3-hydroxybutyrate and acetoacetate. The decrease in the activation of mTORC1 activity in the hypothalamus achieved with ketogenic diets and induced by ketone bodies plays a critical role in blood pressure control, which results in vasodilation and improvements in the blood pressure levels.(187) Therefore, this findings could indicate another possible mechanism through which hesperidin induced blood pressure improvements in subjects with pre- and stage 1 hypertension. Moreover, participants consuming hesperidin also excreted lower urinary amounts of 2-HIB compared to those consuming the CD. This metabolite is also a microbial-derived metabolites resulting from the degradation of dietary proteins.(188) Higher levels of 2-HIB have been identified in the urine of obese people and have been associated with a reduced microbiota diversity. Recently, 2-HIB has also been associated with an increased risk of type 2 diabetes.(189)

In this sense, the present study has a limitation and it is that stool samples were not collected in the CITRUS randomized clinical trial, and the use of these type of samples would have provided more

243 information about metabolites related to microbiota, such as 2-HIB and PAG, and their relationship with the cardiovascular system. In contrast, the subanalysis performed considering the absorption of flavanones is a strength of the study because not all of the subjects have the same characteristics, such as their nutritional status, medication, and microbiota, which could influence the absorption of hesperidin by increasing or decreasing the bioavailability of hesperidin. Thus, whether high hesperidin levels in plasma exert a stronger effect than lower hesperidin levels should be investigated. Moreover, these results highlight the importance of volunteer stratification in clinical trials with hesperidin due to the large interindividual variability in flavanone absorption.

In conclusion, the plasma and urine metabolite hesperetin 7-O-B-D- glucuronide is the only metabolite that differentiated between responses to different hesperidin doses, while urine hesperetin 7-O-B-D- glucuronide metabolite is correlated with low systolic blood pressure levels. In addition, independent of the hesperidin dose, the consumption of hesperidin in OJ and EOJ for 12 weeks has an impact on the serum metabolomic profile by decreasing the levels of endogenous metabolites related to blood pressure (leucine and isoleucine) and inflammation (NAGs). In contrast, hesperidin in OJ and EOJ can also decrease the urinary excretion of uraemic toxins (DMA, NAG and 4- cresuyl sulfate) and endogenous metabolites suggesting an antioxidant effect (pseudouridine). Moreover, 6 h after the consumption of a single dose of hesperidin in OJ and EOJ, changes in the serum levels of metabolites related to reduced blood pressure levels and anti- inflammatory effects (proline betaine, DMG, leucine and isoleucine)

244 were detected. Thus, the nontargeted metabolomics approach offered the possibility of identifying changes in different endogenous metabolites induced by hesperidin consumption that could indicate a beneficial cardiovascular effect of hesperidin and expand our knowledge regarding its potential mechanism of action or biological effects.

5. REFERENCES [1] C. H. Johnson, J. Ivanisevic, G. Siuzdak, Nat. Rev. Mol. Cell

Biol. 2016, 17, 451–459.

[2] D. S. Wishart, Physiol. Rev. 2019, 99, 1819–1875.

[3] Q. N. Van, T. D. Veenstra, H. J. Issaq, Curr. Urol. Rep. 2011,

12, 34–40.

[4] K. Hanhineva, M. A. Lankinen, A. Pedret, U. Schwab, M.

Kolehmainen, J. Paananen, V. de Mello, R. Sola, M. Lehtonen,

K. Poutanen, M. Uusitupa, H. Mykkänen, J. Nutr. 2015, 145,

7–17.

[5] L. Pla-Paga, J. Companys, L. Calderon-Perez, E. Llaurado, R.

Sola, R. M. Valls, A. Pedret, Nutr. Rev. 2019, 77, 1–20.

[6] G. Pereira-Caro, T. Polyviou, I. A. Ludwig, A. M. Nastase, J.

M. Moreno-Rojas, A. L. Garcia, D. Malkova, A. Crozier, Am.

J. Clin. Nutr. 2017, 106, 791–800.

[7] L. Pla-Pagà, M. Guirro, A. Gual-Grau, A. Gibert-Ramos, E.

245

Foguet-Romero, Ú. Catalán, J. Mayneris-Perxachs, N. Canela,

R. M. Valls, L. Arola, R. Solà, A. Pedret, Mol. Nutr. Food Res.

2020, 64, 1–12.

[8] M. Guirro, A. Gual-grau, A. Gibert-ramos, J. M. Alcaide-

hidalgo, N. Canela, J. Mayneris-perxachs, n.d., 1–19.

[9] O. D. Rangel-Huerta, C. M. Aguilera, M. V. Martin, M. J.

Soto, M. C. Rico, F. Vallejo, F. Tomas-Barberan, A. J. Perez-

de-la-Cruz, A. Gil, M. D. Mesa, J. Nutr. 2015, 145, 1808–

1816.

[10] C. Morand, C. Dubray, D. Milenkovic, D. Lioger, J. Franc, A.

Scalbert, Am. J. Clin. Nutr. 2011, 93, 73–80.

[11] N. P. Aptekmann, T. B. Cesar, Maturitas 2010, DOI

10.1016/j.maturitas.2010.07.009.

[12] R. M. Valls, A. Pedret, L. Calderón-Pérez, E. Llauradó, L. Pla-

Pagà, J. Companys, A. Moragas, F. Martín-Luján, Y. Ortega,

M. Giralt, M. Romeu, L. Rubió, J. Mayneris-Perxachs, N.

Canela, F. Puiggrós, A. Caimari, J. M. Del Bas, L. Arola, R.

Solà, Eur. J. Nutr. 2020, DOI 10.1007/s00394-020-02279-0.

[13] C. Manach, C. Morand, A. Gil-Izquierdo, C. Bouteloup-

Demange, C. Rémésy, Eur. J. Clin. Nutr. 2003, 57, 235–242.

[14] F. Vallejo, M. Larrosa, E. Escudero, M. P. Zafrilla, B. Cerdá, J.

246

Boza, M. T. García-Conesa, J. C. Espín, F. A. Tomás-

Barberán, J. Agric. Food Chem. 2010, 58, 6516–6524.

[15] M. Tomás-Navarro, F. Vallejo, E. Sentandreu, J. L. Navarro, F.

A. Tomás-Barberán, J. Agric. Food Chem. 2014, 62, 24–27.

[16] A. Mas-Capdevila, J. Teichenne, C. Domenech-Coca, A.

Caimari, J. M. D. Bas, X. Escoté, A. Crescenti, Nutrients 2020,

12, DOI 10.3390/nu12051488.

[17] M. Yamamoto, H. Jokura, K. Hashizume, H. Ominami, Y.

Shibuya, A. Suzuki, T. Hase, A. Shimotoyodome, Food Funct.

2013, 4, 1346.

[18] A. J. Lloyd, M. Beckmann, G. Favé, J. C. Mathers, J. Draper,

Br. J. Nutr. 2011, 106, 812–824.

[19] M. S. Yoon, Nutrients 2016, 8, DOI 10.3390/nu8070405.

[20] R. Yang, J. Dong, H. Zhao, H. Li, H. Guo, S. Wang, C. Zhang,

S. Wang, M. Wang, S. Yu, W. Chen, PLoS One 2014, 9, DOI

10.1371/journal.pone.0099598.

[21] P. Mirmiran, F. Teymoori, G. Asghari, F. Azizi, Arch. Iran.

Med. 2019, 22, 182–188.

[22] S. C. Deirdre K. Tobias, Patrick R. Lawler, Paulo H. Harada,

Olga V. Demler, Paul M Ridker, JoAnn E. Manson, and S.

Mora, Circ Genom Precis Med 2019, 11, 1–17.

247

[23] C. F.-S. Shannon M. Harlan, Deng-Fu Guo, Donald A.

Morgan, and K. Rahmouni1, Cell Metab. 2013, 17, 599–606.

[24] N. C. B, J. An, J. R. Bain, M. J. Muehlbauer, R. D. Stevens, L.

F. Lien, A. M. Haqq, S. H. Shah, M. Arlotto, C. A. Slentz, D.

Gallup, O. Ilkayeva, B. R. Wenner, W. E. Yancy, G. Musante,

R. Surwit, D. S. Millington, M. D. Butler, P. Laura, Cell

Metab. n.d., 9, 311–326.

[25] M. A. Connelly, J. D. Otvos, I. Shalaurova, M. P. Playford, N.

N. Mehta, J. Transl. Med. 2017, 15, 1–5.

[26] H. Topp, G. Fusch, G. Schöch, C. Fusch, Pediatr. Res. 2008,

64, 246–250.

[27] P. Montenegro Álvarez de Tejera, G. Chamorro Merino, M.

Cabanes Mariscal, P. Sánchez López, J. Medina Font, Sanid.

Mil. 2012, 68, 22–26.

[28] G. Strifler, E. Tuboly, A. Görbe, M. Boros, D. Pécz, P.

Hartmann, PLoS One 2016, 11, 1–14.

[29] and F. E. R. Kymberleigh A. Romano, Ana Martinez-del

Campo, Kazuyuki Kasahara, Carina L. Chittim, Eugenio I.

Vivas, Daniel Amador-Noguez, Emily P. Balskus, Cell Host

Microbe 2017, 22, 279–290.

[30] R. D. Mair, T. L. Sirich, T. W. Meyer, Toxins (Basel). 2018,

248

10, 1–16.

[31] F. Duranton, G. Cohen, R. De Smet, M. Rodriguez, J.

Jankowski, R. Vanholder, A. Argiles, J. Am. Soc. Nephrol.

2012, 23, 1258–1270.

[32] M. Hatem-Vaquero, S. de Frutos, A. Luengo, A. González

Abajo, M. Griera, M. Rodríguez-Puyol, D. Rodríguez-Puyol,

L. Calleros, Nefrología 2018, 38, 639–646.

[33] C. Menni, M. Mangino, M. Cecelja, M. Psatha, M. J. Brosnan,

J. Trimmer, R. P. Mohney, P. Chowienczyk, S. Padmanabhan,

T. D. Spector, A. M. Valdes, J. Hypertens. 2015, 33, 791–796.

[34] F. Ottosson, L. Brunkwall, E. Smith, M. Orho-Melander, P. M.

Nilsson, C. Fernandez, O. Melander, J. Hypertens. 2020,

Publish Ah, DOI 10.1097/hjh.0000000000002569.

[35] I. W. Wu, K. H. Hsu, C. C. Lee, C. Y. Sun, H. J. Hsu, C. J.

Tsai, C. Y. Tzen, Y. C. Wang, C. Y. Lin, M. S. Wu, Nephrol.

Dial. Transplant. 2011, 26, 938–947.

[36] Y. Stevens, E. Van Rymenant, C. Grootaert, J. Van Camp, S.

Possemiers, A. Masclee, D. Jonkers, Nutrients 2019, 11, DOI

10.3390/nu11071464.

[37] L. C. Gormsen, M. Svart, H. H. Thomsen, E. Søndergaard, M.

H. Vendelbo, N. Christensen, L. P. Tolbod, H. J. Harms, R.

249

Nielsen, H. Wiggers, N. Jessen, J. Hansen, H. E. Bøtker, N.

Møller, J. Am. Heart Assoc. 2017, 6, 1–11.

[38] M. Li, B. Wang, M. Zhang, M. Rantalainen, S. Wang, H.

Zhou, Y. Zhang, J. Shen, X. Pang, M. Zhang, H. Wei, Y. Chen,

H. Lu, J. Zuo, M. Su, Y. Qiu, W. Jia, C. Xiao, L. M. Smith, S.

Yang, E. Holmes, H. Tang, G. Zhao, J. K. Nicholson, L. Li, L.

Zhao, Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 2117–2122.

[39] J. Vangipurapu, L. F. Silva, T. Kuulasmaa, U. Smith, M.

Laakso, Diabetes Care 2020, 43, 1319–1325.

ACKNOWLEDGEMENTS

Author contributions The human randomized clinical trial was conducted by LP-P, RMV, AP, LCP, EL, JC, RS, AM, and FM-L. JM-P realized the statistical analysis of metabolomics data. LP-P wrote the manuscript and JM-P, RMV, AP, and RS provided feedback on the manuscript. All the authors have read the manuscript. We thank Nutritional Projects Assessment (Nuproas.es) for their contribution to the statistical analysis.

Funding The Florida Department of Citrus, an executive agency of the state of Florida, USA, provided a grant and study drinks to the Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain. A.P. has Torres Quevedo contract (Subprograma Estatal de Incorporación, Plan Estatal de Investigación Cientíﬁca y Técnica y de Innovación).

250

This work was financially supported by the by the Centre for the Development of Industrial Technology (CDTI) of the Spanish Ministry of Science and Innovation under grant agreement: TECNOMIFOOD project. CER-20191010.

Declaration of interest None of the authors report a conflict of interest related to the study.

251

Table 1. Changes in plasma compliance biomarkers at 4, 8 and 12 weeks after treatments. Treatment Changes among treatments

Control drink OJ EOJ OJ vs Control drink EOJ vs Control drink EOJ vs OJ

Variable Post-int Change Post-int Change Post-int Change Mean P Mean P Mean P (95% CI) (95% CI) (95% CI)

Hesperetin 7-O-B-D glucuronide, nM/L, (log)

Week 4 1.26 ± 0.05 -0.032 1.70 ± 0.09 0.444 1.78 ± 0.566 0.461 <0.001 0.618 <0.001 0.157 0.164 (-0.15;0.08) (0.19;0.70) 0.09 (0.31;0.82) (0.24;0.86) (0.38;0.84) (-0.38;0.06) Week 8 1.23 ± 0.03 -0.031 1.69 ± 0.09 0.443 1.73 ± 0.515 0.451 <0.001 0.542 <0.001 0.090 0.460 (-0.15;0.08) (0.17;0.72) 0.10 (0.22;0.81) (0.21;0.69) (0.30;0.69) (-0.15;0.33) Week 12 1.21 ± 0.02 -0.048 1.68 ± 0.10 0.426 1.88 ± 0.664 0.474 <0.001 0.731 <0.001 0.257 0.040 (-0.19;0.09) (0.14;0.70) 0.11 (0.38;0.95) (0.23;0.71) (0.48;0.98) (0.01;0.50) Hesperetin 3-O-B-D glucuronide, nM/L, (log)

Week 4 0.50 ± 0.00 -0.049 2.33 ± 0.23 0.971 2.59 ± 1.177 1.021 <0.001 1.203 <0.001 0.183 (- 0.281 (-0.18;0.09) (0.59;1.30) 0.22 (0.7;1.60) (0.68;1.30) (0.87;1.50) 0.15;0.52) Week 8 0.54 ± 0.04 -0.06 2.61 ± 0.20 1.249 2.55 ± 1.131 1.257 <0.001 1.040 <0.001 -0.217 0.448 (-0.19; 0.18) (0.50;2.00) 0.23 (0.51;1.70) (0.69;1.0) (0.47;1.60) (-0.78;0.35) Week 12 0.50 ± 0.00 -0.048 2.39 ± 0.22 1.028 2.48 ± 1.059 1.089 <0.001 1.099 <0.001 0.010 0.971 (-0.18; 0.09) (0.41;1.60) 0.25 (0.40;1.70) (0.55;1.60) (0.56;1.60) (-0.52;0.54)

249

Hesperetin 7-O-sulfate, nM/L, (log) Week 4 -0.77 ± -0.188 0.88 ± 0.22 1.257 0.93 ± 1.215 1.537 <0.001 1.685 <0.001 0.148 0.671 0.07 (-0.72;0.34) (0.56;1.90) 0.21 (0.36;2.10) (0.84; (0.99;2.40) (-0.54;0.84) 2.20) Week 8 -0.88 ± -0.302 1.03 ± 0.21 1.404 1.02 ± 1.303 1.738 <0.001 1.689 <0.001 -0.049 0.902 0.00 (-0.77;0.17) (0.45;2.40) 0.22 (0.44;2.20) (0.95;2.50) (0.89;2.50) (-0.84;0.74) Week 12 -0.89 ± -0.302 0.75 ± 0.22 1.122 1.01 ± 1.293 1.485 <0.001 1.679 <0.001 0.194 0.599 0.00 (-0.78;0.17) (0.35;1.90) 0.25 (0.38;2.20) (0.77;2.20) (0.94;2.40) (-0.54;0.92) Naringerin 4-O-B-D glucuronide, nM/L, (log) Week 4 1.22 ± 0.00 -0.074 (- 1.82 ± 0.12 0.490 1.52 ± 0.259 0.516 <0.001 0.420 0.004 -0.096 0.493 0.22;0.07) (0.14;0.83) 0.09 (-0.005;0.52) (0.24;0.79) (0.14;0.70) (-0.18;0.37) Week 8 1.21 ± 0.03 -0.073 (- 1.66 ± 0.11 0.335 1.59 ± 0.330 0.386 0.012 0.414 0.008 0.028 0.855 0.14;0.08) (-0.04;0.71) 0.12 (-0.009;0.67) (0.08;0.69) (0.11;0.72) (-0.27;0.33) Week 12 1.22 ± 0.00 -0.074 (- 1.62 ± 0.11 0.296 1.71 ± 0.457 0.412 0.007 0.555 0.001 0.142 0.353 0.22;0.07) (-0.03;0.62) 0.13 (0.08;0.83) (0.11;0.71) (0.25;0.86) (-0.16;0.44) Naringerin glucuronide, nM/L, (log)

Week 4 1.24 ± 0.00 -0.074 1.77 ± 0.09 0.443 1.66 ± 0.392 0.511 <0.001 0.559 <0.001 0.048 0.682 (-0.22;0.07) (0.18;0.71) 0.09 (0.12;0.66) (0.28;0.74) (0.32;0.79) (-0.18;0.28) Week 8 1.21 ± 0.03 -0.073 1.75 ± 0.10 0.428 1.59 ± 0.322 0.509 <0.001 0.431 0.002 -0.079 0.561 (-0.14;0.08) (0.09;0.77) 0.10 (0.02;0.62) (0.24;0.78) (0.16;0.70) (-0.35;0.19)

250

Week 12 1.22 ± 0.00 -0.074 1.70 ± 0.09 0.380 1.75 ± 0.485 0.488 <0.001 0.586 <0.001 0.098 0.429 (-0.22;0.07) (0.11;0.65) 0.09 (0.19;0.78) (0.25;0.73) (0.34;0.83) (-0.15;0.34) Naringerin sulfate, nM/L, (log) Week 4 -0.81 ± -0.101 -0.17 ± 0.549 -0.35 ± 0.418 0.640 0.001 0.660 <0.001 0.019 0.914 0.04 (-0.36;0.16) 0.15 (0.15;0.95) 0.16 (0.04;0.79) (0.28;1.00) (0.30;1.00) (-0.33;0.37) Week 8 -0.79 ± -0.081 -0.24 ± 0.486 -0.38 ± 0.387 0.577 0.006 0.526 0.014 -0.051 0.807 0.06 (-0.32;0.16) 0.17 (-0.06;1.00) 0.16 (-0.01;0.79) (0.16;0.99) (0.11;0.94) (-0.46;0.36) Week 12 -0.85 ± -0.140 -0.30 ± 0.417 -0.25 ± 0.521 0.635 0.001 0.696 <0.001 0.061 0.739 0.00 (-0.37;0.10) 0.15 (0.04;0.79) 0.18 (0.09;0.95) (0.28;0.99) (0.33;1.10) (-0.30;0.42) Data expressed as mean ± standard error or mean (95% confidence interval, CI). Intra treatment comparisons evaluated by General Lineal Model with Bonferroni correction and age and sex as covariables. Inter treatment comparisons by ANCOVA Model adjusted by sex and age. P<0.05. Abbreviations: Post-in, post-treatment values; Change, change from baseline; OJ, orange juice; EOJ, hesperidin-enriched orange juice.

251

252

Figure 1. Targeted metabolomics results in plasma samples after the single dose consumption of control drink, orange juice and hesperidin-enriched orange juice

(N=52). P<0.05; **, P<0.01; **, P<0.001; ns, nonsignificant.

253

Table 2. Changes in urine of compliance biomarkers after sustained consumption of treatments (N=129). Treatment Changes among treatments CD OJ EOJ OJ vs CD EOJ vs CD EOJ vs OJ Change Post-int Change Post-int Change Post-int Median P Median P Median P (25th-75th p) (25th-75th p) (25th-75th p) Hesperetin 7-O-β-D glucuronide, nM 0.00 0.00 0.00‡ 0.00 116‡ 116 0.00 <0.001 116 <0.001 116 0.030

(0.00; 0.00) (0.00; 0.00) (0.00; 137) (0.00; 137) (0.00; 347) (0.00; 347) (0.00;137 (0.00;137 (0.85;210)

Hesperetin 3-O-β-D glucuronide, nM 0.00 0.00 495 ‡ 1493 1281 ‡ 1499 495 <0.001 1281 <0.001 786 0.265

(0.00; 0.00) (0.00; 0.00) (0.00; (0.00; (0.00; 6344) (0.00; (0.00;2234) (0.00; (0.00;4110)

2234) 2870) 6344) 6344)

Hesperetin 7-O-sulfate, nM

0.00 0.00 2.16† 24 43‡ 43 2.16 <0.001 43 <0.001 41 0.136

(0.00; 0.00) (0.00; 0.00) (0.00; 90) (0.00; 119) (0.00; 209) (0.00; 218) (0.00; 90) (0.00; 209) (0.00;119)

254

Naringerin 4-O-β-D glucuronide, nM 0.00 0.00 0.00 0.00 0.00† 0.00 0.00 0.004 0.00 <0.001 0.00 0.569

(0.00; 0.00) (0.00; 0.00) (0.00; 0.00) (0.00; 140) (0.00; 308) (0.00; 308) (0.00; 0.00) (0.00; 308) (0.00; 308)

Naringerin glucuronide, nM 0.00 0.00 0.00† 0.00 0.00† 0.00 0.00 <0.001 0.00 <0.001 0.00 0.413

(0.00; 0.00) (0.00; 0.00) (0.00; 205) (0.00; 232) (0.00; 195) (0.00; 195) (0.00; 205) (0.00; 195) (-5;0.00)

Naringerin sulfate, nM 0.00 0.00 0.00* 0.00 0.00† 0.00 0.00 0.001 .00 0.001 0.00 0.601 (0.00; 0.00) (0.00; 0.00) (0.00; 0.11) (0.00; 1.08) (0.00; 3.12) (0.00; 3.12) (0.00; 0.11) (0.00; 3.12) (0.00; 3.01) j Data expressed as median (25th-75th percentiles) /median (minimum-maximum) Intra-treatment comparisons by Wilcoxon test, * P<0.05; †P<0.005, ‡P<0.001. Inter- treatment comparisons by Man-Whitney test for independent samples. Abbreviations: CD, control drink; OJ, orange juice; EOJ, hesperidin-enriched orange juice.

255

Figure 2. Relationship between levels of hesperetin-7-O-β-d glucuronide in urine and changes in systolic blood pressure after 12 weeks of treatments.

256

Figure 3. O-PLS coefficients plots for the nontargeted plasma metabolomics analyses after 6-h of beverage consumption in the single dose study (N=52). (A) Orange juice vs control drink. (B) Hesperidin-enriched orange juice vs CD. (C) Maximal total flavanone concentration during the single dose study for each participant in the three treatment groups. (D) O-PLS coefficients plot considering the maximal total flavanone concentration as the response variable. Abbreviations: OJ, orange juice; DMG, dimethylglycine; EOJ, hesperidin-enriched orange juice.

257

Figure 4. O-PLS-DA coefficient plots for the plasma nontargeted metabolomics analyses at the end of the chronic study (N=52). (A) Orange juice vs Control Drink. (B) Hesperidin- enriched orange juice vs Control Drink. Abbreviations: OJ, orange juice; GPC, glycerophosphocholine; NAG, N-acetylglycoproteins; EOJ, hesperidin-enriched orange juice.

258

Figure 5. O-PLS-DA coefficient plots for the urine nontargeted metabolomics analyses at the end of the chronic study (N=129). (A) Orange juice vs Control Drink. (B) Hesperidin- enriched orange juice vs Control Drink. Abbreviations: OJ, orange juice; PAG, phenylacetylglutamine; DMA, dimethylamine; NAG, N-acetylglycoproteins; EOJ, hesperidin- enriched orange juice.

259

Supplementary Information Methods S1

Study population: Participants were recruited between January 2016 and June 2017 thought social networks and newspapers in the Hospital Universitari Sant Joan of Reus (Spain). The inclusion criteria were adults from 18 years to 65 years, with systolic blood pressure level ≥ 120 mm Hg and no family history of CVDs or other chronic diseases. The exclusion criteria were subjects with: body mass index ≥35 kg/m2, fasting glucose >125 mg/dL, hemoglobin ≤13 mg/dL in men and ≤12 mg/dL in women, systolic blood pressure ≥160 mm Hg and diastolic blood pressure >100 mm Hg or taking anti-hypertensive medications, smoking, alcoholism, pregnancy, use of antioxidants or vitamin supplements, and following a vegetarian diet.

All the volunteers signed the informed consent before their inclusion in the clinical trial. The randomized clinical trial received the approbation by the Clinical Research Ethical Committee of Hospital Universitari Sant Joan of Reus (14-12-18/12aclaassN1). The study was conducted in accordance with Helsinki Declaration and Good Clinical Practice Guidelines of the International Conference of Harmonization and was registered at Clinical-Trials.gov: NCT02479568.

Study design: The study was a randomized, parallel, double-blind, and placebo- controlled clinical trial. The subjects were randomly assigned by a computerized random-number generator to 1 group of the 3 intervention groups: control drink (CD), orange juice (OJ) and hesperidin-enriched

260 orange juice (EOJ). All the participants consume daily 500 mL of the corresponding test drink for 12 weeks, and 2 dose-responses studies were performed at baseline and after 12 weeks. Only in the baseline dose-response study were collected plasma samples for targeted and non-targeted metabolomics approach.

261

Supplemental Information Figure S2. Total absorption of flavanones in baseline conditions and at 2, 4 and 6 hours after each treatment. Abbreviations: OJ, orange juice; EOJ, hesperidin-enriched orange juice.

262

Supporting Information Table S1. Composition of the 3 test drinks. Test drink (for 500 mL/day)

Control drink Orange juice Hesperidin-enriched orange juice Acidity, % 2.49 2.49 2.49

Sugar, g 43.1 37.7 37.7

Vitamin C, mg 235.3 235.3 235.3

Citric acid, g 3.40 3.40 3.40

Narirutin, mg ND 64 77

Hesperidin, mg ND 392 670

The 3 test drinks were in frozen concentrated canned drink, once diluted 3.4:1 (water to syrup). Abbreviations: ND, non-detectable.

263

Supplementary Information Table S2. Results from the method validation for targeted metabolomics by UHPLC-MS/MS.

Metabolites Abbreviation MDL MQL Repeatability (µM) (µM) (RSD) (n=20) (%) Hesperetin 7-O-beta-D- H7G 20.9 62.8 10.8 glucuronide Hesperetin 3-O-beta-D- H3G 85.3 255.9 6.7 glucuronide Hesperetin 7-O-sulfate H7S 0.2 0.45 4.6

Naringenin 4'-O-beta-D- N4G 51.4 154.1 10.1 glucuronide Naringenin-glucuronide NG - - 7.5

Naringenin-sulfate N-S - - 4.5

264

Supplementary Information Table S3. Baseline characteristics of participants by intervention group. Variable CD (n=53) OJ (n=53) EOJ (n=53) P Age, y 45.4 ±13.0 43.3 ± 12.0 43.6 ± 11.8 0.629

Females, % 34.0 32.1 34.0 0.981

SBP, mm Hg 132 ± 9.94 132 ± 9.11 134 ± 9.82 0.687

DPB, mm Hg 79 ± 8.14 80 ± 8.42 79 ± 10.2 0.868

Pulse pressure, mm Hg 53 ± 9.09 52 ± 8.05 54 ± 6.74 0.261

Weight, kg 77.3 ± 15.4 78.8 ± 12.2 75.9 ± 11.6 0.523

BMI, kg/m2 26.1 ± 3.8 26.4 ± 3.6 26.1 ± 3.3 0.858

Waist circumference, cm 93.0 ± 11.0 91.7 ± 10.9 91.4 ± 10.7 0.766

Waist/height, cm 0.54 ± 0.06 0.53 ± 0.07 0.54 ± 0.07 0.790

Conicity index 1.50 ± 0.76 1.30 ± 0.35 1.39 ± 0.62 0.269

Glucose, mg/dL 91.6 ± 9.2 93.6 ± 11.6 93.6 ± 9.6 0.517

Cholesterol, mg/dL

Total 196 ± 30.1 198 ± 32.7 196 ± 31.6 0.937

LDL 124 ± 26.4 125 ± 31.5 127 ± 25.1 0.900

HDL 50.9 ±13.4 51.0 ± 14.7 49.8 ± 13.0 0.889

Triglycerides*, mg/dL 82 (67-118) 85 (65-121) 81 (63-116) 0.624

Physical activity, A 3.08 ± 0.06 3.12 ± 1.38 3.12 ± 1.26 0.986 Data are expressed as the mean ± standard deviation, or percentages. CD, control drink; OJ, orange juice; EOJ, hesperidin- enriched orange juice; SBP, systolic blood pressure; DBP, diastolic blood pressure; Pulse pressure = SBP-DBP; BMI, body mass index; LDL, low-density lipoproteins; HDL, high-density lipoproteins * median (25th -75th percentiles). AU, arbitrary units: 0, inactive; 1, very low activity; 2, low activity; 3, moderately active; 4, very active. P for ANOVA with logarithmic transformation for triglycerides.

265

Project 5: Proteomic analysis of heart and kidney tissues in healthy and metabolic syndrome rats after hesperidin supplementation.

266

267

RESEARCH ARTICLE

www.mnf-journal.com Proteomic Analysis of Heart and Kidney Tissues in Healthy and Metabolic Syndrome Rats after Hesperidin Supplementation

Laura Pla-Pagà, Maria Guirro, Andreu Gual-Grau, Albert Gibert-Ramos, Elisabet Foguet-Romero, Úrsula Catalán, Jordi Mayneris-Perxachs,* Nuria Canela, Rosa M. Valls,* Lluís Arola, Rosa Solà, and Anna Pedret

1. Introduction Scope: Proteomics has provided new strategies to elucidate the mechanistic action of hesperidin, a flavonoid present in citrus fruits. Thus, the aim of the Diets rich in fruits and vegetables are known to protect against cardiovascu- present study is to determine the effects of hesperidin supplementation (HS) lar diseases (CVDs),[1,2] which include on the proteomic profiles of heart and kidney tissue samples from healthy and heart failure, stroke, and chronic kidney metabolic syndrome (MS) rats. diseases and are the leading cause of mor- Methods and results: 24 Sprague Dawley rats are randomized into four tality worldwide.[3] Several studies have groups: healthy rats fed with a standard diet without HS, healthy rats aimed to elucidate the beneficial role of − − bioactive compounds present in food, administered with HS (100 mg kg 1 day 1), MS rats without HS, and MS rats −1 −1 such as phenolic compounds, which have administered with HS (100 mg kg day ) for eight weeks. Heart and kidney shown beneficial effects on different car- samples are obtained, and proteomic analysis is performed by mass diovascular risk factors (CVRFs)[4] and on spectrometry. Multivariate, univariate, and ingenuity pathways analyses are the prevention of CVDs,[5] to determine performed. Comparative and semiquantitative proteomic analyses of heart their mechanisms of action and iden- and kidney tissues reveal differential protein expression between MS rats with tify biomarkers of disease or treatment response. In this sense, omics sciences and without HS. The top diseases and functions implicated are related to the have gained attention since they can cardiovascular system, free radical scavenging, lipid metabolism, glucose provide important biological information metabolism, and renal and urological diseases. on many biomolecules. Proteomics, one Conclusion: This study is the first to demonstrate the protective capacity of of the most common omics sciences hesperidin to change to the proteomic profiles in relation to different involves large-scale protein identification to study the proteome of a tissue cardiovascular risk biomarkers in the heart and kidney tissues of MS rats. or organ under certain conditions. In

L. Pla-Pagà, Dr. R. M. Valls, Prof. L. Arola, Prof. R. Solà, Dr. A. Pedret Dr. M. Guirro, A. Gual-Grau, Dr. A. Gibert-Ramos, Prof. L. Arola Eurecat Nutrigenomics Research Group Centre Tecnològic de Catalunya Biochemistry and Biotechnology Department Unitat de Nutrició i Salut Universitat Rovira i Virgili Av/ Universitat 1, Reus 43204, Spain C/ Marcel·lí Domingo 1, Tarragona 43007, Spain L. Pla-Pagà, Dr. Ú. Catalán, Dr. R. M. Valls, Prof. R. Solà, Dr. A. Pedret Dr. M. Guirro, E. Foguet-Romero, Dr. J. Mayneris-Perxachs Universitat Rovira i Virgili Centre for Omic Sciences Facultat de Medicina i Ciències de la Salut Joint Unit Universitat Rovira i Virgili—EURECAT Functional Nutrition Centre Tecnològic de Catalunya Oxidation and Cardiovascular Disease Group (NFOC-SALUT) Unique Scientiﬁc and Technical Infrastructures C/ Sant Llorenç 21, Reus 43201, Spain Av/ Universitat 1, Reus 43204, Spain E-mail: [email protected] E-mail: [email protected] Dr. Ú. Catalán, Dr. N. Canela Institut d’Investigació Sanitària Pere Virgili Av/ Universitat 1, Reus 43204, Spain The ORCID identiﬁcation number(s) for the author(s) of this article Prof. R. Solà can be found under https://doi.org/10.1002/mnfr.201901063 Hospital Universitari Sant Joan Av/ Doctor Josep Laporte 2, Reus 43204, Spain DOI: 10.1002/mnfr.201901063

Mol. Nutr. Food Res. 2020, 64, 1901063 1901063 (1 of 12) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.mnf-journal.com cardiovascular research, proteomics can help to elucidate the sig- 2.2. Dosage Information naling mechanisms involved in CVDs[6] and facilitate prevention and drug development.[7] Hesperidin was administered daily and orally via low-fat con- − In recent years, citrus, particularly orange and orange juice, densed milk over eight weeks at a dose of 100 mg kg 1 of body has been investigated in cardiovascular research. Citrus is weight per day. The rationale for choosing this dose was based on rich in polyphenols, mainly flavonoids, and lignans, with hes- the beneficial effects of hesperidin reported in previous works fo- peridin (hesperetin-7-O-rutinoside) being the most abundant. cused on MS factors in rat models.[14,22] Accordingly, the human − Hesperidin has been assessed in studies on rats and humans and equivalent dose of 100 mg kg 1 hesperidin was 1350 mg per day shown to have beneficial effects on several parameters related for a 60 kg human,[21,23] a dose achievable with hesperidin-rich to the cardiovascular system and the improvement of CVRFs, orange juice. such as decreasing blood pressure,[8–11] improving endothelium- dependent vasodilation during hypertension,[12] decreasing total cholesterol and triglyceride levels,[13] improving glucose and insulin levels and the homeostasis model assessment index,[14] 2.3. Kidney and Heart Proteomic Analysis decreasing inflammatory markers,[9] decreasing kidney damage markers,[15] and decreasing oxidative stress.[16] Heart and kidney tissue samples were obtained immediately after However, proteomic analyses of the effects of citrus on health the animals were sacrificed, snap-frozen in liquid nitrogen, and are scarce. Some studies analyzed the orange proteome but only stored at −80 °C until the analyses were performed. for technological purposes.[17,18] To the best of our knowledge, only one study performed an interventional analysis of orange juice consumption in humans to evaluate the proteomic changes in the peripheral blood mononuclear cells of healthy humans after consumption of a high-fat and high-carbohydrate meal to 2.3.1. Protein Extraction and Quantification evaluate oxidative stress and inflammatory markers.[19] In this study, orange juice suppressed diet-induced inflammation. Fur- Sample tissue was weighed (25–30 mg) to perform cell lysis, re- thermore, only one study evaluated the effects of hesperidin on alized according to the radioimmunoprecipitation assay buffer the proteomic profile of human HepG2 cells in relation to cell (RIPA) protocol (ThermoFisher Scientific, Barcelona, Spain). [20] death. Therefore, no studies have examined the effects of hes- First, the samples were frozen in liquid nitrogen for complete peridin on the human or rat tissue proteome in relation to CVD or lysis. The samples were then mixed with 1 mL of RIPA buffer, ho- CVRFs despite several publications reporting a beneficial effect of mogenized completely with a BlueBender via freeze thaw cycles, hesperidin. Furthermore, the biological processes by which hes- agitated for 1 h at 4 °C, and centrifuged. After centrifugation, the peridin can induce cardioprotective effects have not been eluci- samples were sonicated with a 30 s pulse at 50% amplitude. The dated through proteomic analyses. samples were then centrifuged at 21 130 relative centrifugal force Therefore, the aim of this work was to determine the changes for 15 min, and the supernatants were collected for protein pre- in the proteomic profiles of heart and kidney tissues in healthy cipitation with the addition of 10% trichloroacetic acid/acetone. and metabolic syndrome (MS) rats after hesperidin supplemen- The protein pellets were resuspended in 6 m urea per 50 mm am- tation to shed light on the hesperidin mechanism of action. monium bicarbonate and quantified by Bradford’s method.

2. Experimental Section 2.3.2. Protein Digestion and Peptide 10-Plex Tandem Mass Tag 2.1. Animals and Experimental Design Labeling

The animal procedures were conducted in accordance with the Thirty micrograms of total protein was reduced with 4 mm 1.4- guidelines of the EU Directive 2010/63/EU for animal experi- dithiothreitol for 1 h at 37 °C and alkylated with 8 mm iodoac- ments and approved by the Government of Catalonia and the etamide for 30 min at 25 °C in the dark. Afterward, the sam- Animal Ethics Committee of the University Rovira i Virgili (num- ples were digested overnight (pH 8, 37 °C) with sequencing-grade ber 10 061). 24 eight-week-old male Sprague Dawley rats (Charles trypsin (Promega, Wisconsin, USA) at an enzyme:protein ratio of River Laboratories, Wilmington, Massachusetts, USA) were ran- 1:50. Digestion was quenched by acidification with 1% v/v formic domly assigned to one of the following four groups (n = 6per acid, and peptides were desalted on an Oasis HLB SPE column group): healthy rats fed a standard diet and supplemented with (Waters, California, USA) before TMT 10-plex labeling (Thermo milk as a vehicle (STDV), healthy rats fed a standard diet and Fisher, Scientific, Massachusetts, USA) according to the manu- supplemented with milk containing hesperidin (STDH), MS-rats facturer’s instructions. fed a cafeteria diet supplemented with milk as a vehicle (CAFV), For normalization of all samples in the study along with the dif- and MS-rats fed a cafeteria diet supplemented with milk contain- ferent TMT-multiplexed batches, a pool containing all the sam- ing hesperidin (CAFH); all of the treatments occurred over eight ples was labeled with a TMT-126 tag and included in each TMT weeks. The experimental design was the same as that described batch. The different TMT 10-plex batches were desalted on Oasis by Guirro M et al.[21] HLB SPE columns before nanoLC-MS analysis.

2.3.3. Off Gel-NanoLC-(Orbitrap) MS/MS Analysis univariate analysis with a p-value < 0.01 and a PLS-DA variable importance on projection (VIP) score > 1.5 in multivariate anal- Labeled and multiplexed samples were fractionated by off-gel ysis were considered differentially expressed between groups. technology (Agilent, California, USA) according to the manufacturer’s protocol. Samples were fractioned on 12 cm non-linear pH 3–10 strips in 12 fractions. Thus, fraction 1 (F1) was mixed 2.5. Pathway Analysis with fraction 7 (F7), F2 was mixed with F8, and this protocol was repeated with all fractions. In total, 6 fractions were ob- The UniProt Database was used to obtain the gene symbols tained, and they were separated on a C-18 reversed-phase (RP) and protein description (https://www.uniprot.org/). Ingenuity nanocolumn (75 𝜇mI.D;15cmlength;3𝜇m particle diameter, pathway analysis (IPA; www.ingenuity.com) was used to ana- Nikkyo Technos Co., Ltd., Japan) on an EASY-II nanoLC made by lyze the protein networks via the statistically significant results Thermo Fisher Scientific (Massachusetts, USA). The chromato- from both multivariate and univariate analyses for biological in- graphic separation was performed with a 90-min gradient using terpretation. IPA was used to explore the possible metabolic cell Milli-Q water (0.10% formic acid) and acetonitrile (0.10% formic signaling pathways that were over- or underrepresented by the acid) as the mobile phase at a flow rate of 300 nL min−1. experimentally-determined proteins. Mass spectrometry analyses were performed on an LTQ- Orbitrap Velos Pro made by Thermo Fisher via an enhanced 3. Results FT-resolution MS spectrum (R = 30 000 FHMW) followed by a data-dependent FT-MS/MS acquisition (R = 15 000 FHMW, 40% 3.1. Biochemical Analyses HCD) of the 10 most intense parent ions with a charge state re- jection of one and a dynamic exclusion of 0.5 min. In previous studies,[21,24] the feeding of a cafeteria diet to Sprague Dawley rats induced an obesogenic pattern with significant increases in body weight and fat mass, elevated systolic blood 2.3.4. Protein Identification and Quantification pressure, hypertriglyceridemia, hyperglycemia, and high levels of low-density lipoprotein cholesterol (LDL-c), which are crite- Protein identification and quantification were performed with ria required for MS diagnosis. Consequently, when applying the Proteome Discoverer software v.1.4.0.288 (Thermo Fisher Sci- harmonized human definition of MS to Sprague Dawley rats entific, Massachusetts, USA) using multidimensional protein consuming a cafeteria diet, the rats exhibited a reflex of fat identification technology, combining the six raw data files ob- mass, elevated triglyceride (TG) levels (drug treatment for ele- tained after off-gel fractionation. For protein identification, all MS vated TG is an alternate indicator), reduced high-density lipopro- and MS/MS spectra were analyzed using the Mascot search en- tein cholesterol (HDL-c) levels (drug treatment for reduced HDL- gine (v.2.5). Mascot was set up to search SwissProt_2018_03.fasta c is an alternate indicator), elevated blood pressure (systolic or database (557 012 entries), restricted to Rattus norvegicus tax- diastolic (antihypertensive drug treatment in a person with a his- onomy (8003 sequences) and assuming trypsin digestion. Two tory of hypertension is an alternate indicator), and/or elevated missed cleavages were allowed, and errors of 0.02 Da for an fasting plasma glucose levels (drug treatment of elevated glucose FT-MS/MS fragmentation mass and 10 ppm for an FT-MS par- is an alternate indicator).[25] ent ion mass were allowed. TMT-10plex was set as the quan- We also report that the consumption of hesperidin (100 mg tification modification, oxidation of methionine and acetyla- kg−1 body weight per day) for eight weeks improved lipid tion of N-termini were set as dynamic modifications, and car- metabolism and the insulin response and decreased the systolic bamidomethylation of cysteine was set as the static modification. blood pressure in MS rats. In this sense, hesperidin supplemen- For protein quantification, the ratios between each TMT-label tation can improve most of the MS criteria.[24] and the 126-TMT label were used, and quantification results were normalized based on the protein median. 3.2. Proteomic Analysis Results 2.4. Statistical Analysis From the proteomic analysis of heart and kidney tissue rat sam- The proteins present in ≥67% of the samples in the four groups ples, 1127 and 1753 proteins were identified, respectively, and the were considered for the statistical analysis. After the proteomic total proteins identified from each tissue are detailed in Tables analysis, the data were log base 2 transformed, mean centered, S1 and S2, Supporting Information. The tables contain informa- and Pareto scaled. The multivariate statistical analysis was per- tion on the protein IDs from UniProt, descriptions of the pro- formed using Metaboanalyst 4.0 (https://www.metaboanalyst. teins, coverage, unique peptides identified, total peptides iden- ca/). The modeling included the use of unsupervised methods tified, peptide-to-spectrum matches, molecular weights, and the such as principal component analysis (PCA), supervised meth- value for each protein. ods such as partial least squares discriminant analysis (PLS-DA), and an orthogonal projection to latent structures discriminant analysis (OPLS-DA). For the univariate statistical analysis, the 3.2.1. Proteomic Analysis of Heart Rat Tissue distribution of normality was assessed by Kolmogorov-Smirnov tests, and a t-test or a Wilcoxon test was performed for pairwise For statistical analyses only those proteins that were present comparisons. The proteins that were statistically significant in in ≥67% of the samples in the four groups were considered.

After filtering, a total of 872 proteins were considered for statis- eight weeks were: APOA4, SOD1, ATP5F1D, ATPA1A, SLC27A1, tical analysis. The identified proteins with a p-value < 0.01 and a SLC25A3, OGDH, FLNA, ALDH7A1, CLTC, and CDC42. The VIP score from PLS-DA > 1.5 were considered significantly dif- other proteins implicated in the top networks related to the pro- ferent between the STDV and STDH groups, and between the teins analyzed in the present study were: IL-1B, LAMC1, TNF, CAFV and CAFH groups. KCNJ11, ATP1B1, OTOF, SLC1A2, AP2B1, EPB41L2, GRM4, Hesperidin Effects in Healthy Rats Fed with a Standard Diet:In CACNA1B, F11R, TNFRSF1B, KCNJ2, CPT1A, PIN, MYO1C, multivariate analysis, no clear differences were observed between PP2A, PRKCD, STAT3, NEFH, ITSN1, SRC, ABR, PRKCZ, the two STD treatment groups (STDV vs STDH) in the unsuper- CAMKII, FILIP1, AR, RCAN1, GSK3B, NFKB complex, and vised analysis. The supervised analysis showed that none of the SIPI. assessed models was significant due to the negative performance measurement Q2, indicating that the models were not predictive at all or were overfitted and that a random model would perform 3.2.2. Proteomic Analysis of Kidney Rat Tissue better. Moreover, the univariate analysis showed no significant differences between the treatment groups (STDV vs STDH) in After filtering, a total of 1341 proteins were considered for any of the 872 considered proteins. the statistical analysis of kidney tissue samples. The identi- Hesperidin Effects in MS Rats Fed with a Cafeteria Diet:Inmul- fied proteins with a p-value < 0.01 and a VIP score from PLS- tivariate analysis, no clear differences between the two treatment DA > 1.5 were considered significantly different between the groups (CAFV vs CAFH) were observed in CAF rats from the un- STDV and STDH groups, and between the CAFV and CAFH supervised analysis. The PLS-DA from the supervised analysis groups. showed that a model including one component provided the best Hesperidin Effects in Healthy Rats Fed a Standard Diet:From performance, as determined by the Q2 measure. However, after the multivariate analysis no clear separation between the two permutation testing, the model was not found to be significant. STD groups (STDV versus STDH) was evident from the un- Nevertheless, a borderline significant model with a strong pre- supervised or supervised analysis. From the univariate anal- dictive ability (Q2Y = 0.58, p = 0.053) was obtained for the com- ysis, no statistically significant proteins were observed be- parison between CAFH and CAFV rats. The univariate analysis tween the two STD groups among any of the 872 considered showed 65 differentially expressed proteins between CAF groups proteins. with a p-value < 0.01. Hesperidin Effects in MS Rats Fed a Cafeteria Diet:Fromun- The proteins with a p-value < 0.01 and a VIP score from PLS- supervised analysis, a clear separation between the two CAF DA > 1.5 were considered to be differentially expressed between groups (CAFV versus CAFH) was evident from the PCA, al- the group without hesperidin supplementation (CAFV) and the though it was not very clear in the hierarchical clustering anal- group with hesperidin supplementation (CAFH). A total of 35 ysis. A PLS-DA from the supervised analysis showed that a proteins considered to be significantly different between the two model including one component provided the best performance, groups met the two criteria of univariate and multivariate analy- as determined by the Q2 measurement. The OPLS-DA analyses. The information about the 35 proteins is detailed in Table 1. sis showed a significant model with a strong predictive abil- In total, 19 proteins were downregulated and 16 were upregulated ity (Q2Y = 0.70, p = 0.046) for the comparison between CAFH after hesperidin treatment for eight weeks compared to those in and CAFV rats. The univariate analysis showed 75 proteins that the CAFV group. were differentially expressed between the two CAF groups with a Pathway Analysis of the Heart Tissue Proteome: IPA analysis p-value < 0.01. was performed and the top signaling pathways that were sig- The proteins with a p-value < 0.01 and a VIP score from PLS- nificantly affected after hesperidin supplementation were ob- DA > 1.5 were considered to be differentially expressed between tained. The top five significant signaling pathways were: produc- the groups with and without hesperidin supplementation. A to- tion of nitric oxide and reactive oxygen species in macrophages tal of 53 proteins were considered to be significantly different be- (p = 4.08E−04; ratio = 0.021), which involved: APOA4, APOC2, tween the two groups met the two criteria from univariate and CDC42, and RBP4; clathrin-mediated endocytosis signalling multivariate analyses. The information on the 53 proteins is de- (p = 4.50E−04; ratio = 0021), which involved: APOA4, APOC2, tailed in Table 2. In total, 33 proteins were downregulated and CLTC, and RBP4; LXR/RXR activation (p = 1.07E−03; ratio = 20 were upregulated after hesperidin treatment for eight weeks 0.028), atherosclerosis signaling (p = 1.19E−03; ratio = 0.027); compared to those in the CAFV group. and FXR/RXR activation (p = 1.25E−03; ratio = 0.026) which in- Pathway Analysis of the Kidney Tissue Proteome: IPA analysis volved: APOA4, APOC2, and RBP4 in the last three canonical was performed and the top signaling pathways that were signifi- pathways. cantly affected in kidney tissue after hesperidin supplementation Network of the Heart Tissue Proteome: The top networks found were obtained. The top five significant signaling pathways were: by IPA were “Molecular transport, carbohydrate metabolism, and mitochondrial dysfunction (p = 4.65E−06; ratio = 0.039), which small molecule biochemistry” (score = 16), “Cardiac arrhyth- involved ATP5PF, COX6B1, CPT1A, OGDH, TXN2, and VDAC3; mia, cardiovascular disease, and metabolic disease” network the sirtuin signalling pathway (p = 9.37E−05; ratio = 0.023), (score = 2), and “Cancer, cardiovascular disease, and cell cycle” which involved: ATP5PF, CPT1A, HIST2H3C, SOD1, TIMM8B, (score = 2). The graphical representation of the top overlapping and VDAC3; xanthine and xanthosine salvage (p = 2.82E−03; networks and the up- and downregulated proteins, symbolized ratio = 1), guanine and guanosine salvage I (p = 5.62E−03; in red or green, respectively, is shown in Figure 1.Theproteins ratio = 0.5); and adenine and adenosine salvage I (p = 5.62E−03; implicated in the top networks after hesperidin treatment for ratio = 0.5), which involved PNP.

Table 1. 35 proteins diﬀerentially expressed in heart tissue after hesperidin treatment in CAFH compared to CAFV rats.

UniProt code Gene symbol Protein description MW [kDa] Univariate analysis Multivariate analysis

FC p-valuea) VIPb)

M0RDK9 ACAD8 Acyl-CoA dehydrogenase family, member 8 45.10 1.4409 0.0044 1.8469 F1LN92 AFG3L2 AFG3-like matrix AAA peptidase subunit 2 89.30 1.3967 0.0060 1.7448 Q64057 ALDH7A1 Alpha-aminoadipic semialdehyde dehydrogenase 58.70 1.2888 0.0004 1.6283 P02651 APO A4 Apolipoprotein A-IV 44.40 −1.5095 0.0093 1.9047 P19939 APO C1 Apolipoprotein C-I 9.90 1.7180 0.0033 2.2661 G3V8D4 APO C2 Apolipoprotein C-II 10.70 −1.8310 0.0006 2.4926 P06685 ATP1A1 Sodium/potassium-transporting ATPase subunit alpha-1 113.00 1.3297 0.0020 1.6668 G3V7Y3 ATP5F1D ATP synthase subunit delta, mitochondrial 17.60 −1.3813 0.0047 1.7317 D4A305 CCDC58 Coiled-coil domain containing 58, isoform CRA_c 16.70 1.7327 0.0065 2.2324 Q8CFN2 CDC42 Cell division control protein 42 homolog 21.20 1.2746 0.0025 1.5289 Q5XIM5 CDV3 Protein CDV3 homolog 24.30 −1.6055 0.0040 2.1074 M0RC65 CFL2 Coﬁlin 2 18.70 −1.3380 0.0054 1.635 F1M779 CLTC Clathrin heavy chain 191.40 1.2960 0.0024 1.5829 Q5BJQ0 COQ8A Atypical kinase COQ8A, mitochondrial 72.20 −1.2667 0.0012 1.5373 P11240 COX5A Cytochrome c oxidase subunit 5A, mitochondrial 16.10 −1.5030 0.0056 1.9324 P60841 ENSA Alpha-endosulﬁne NA −1.2770 0.0034 1.5225 C0JPT7 FLNA Filamin A 280.30 1.5440 0.0007 2.1071 D3ZT90 GCDH Glutaryl-CoA dehydrogenase 49.70 1.2996 0.0050 1.5552 Q5I0P2 GCSH Glycine cleavage system H protein, mitochondrial 18.50 −1.5136 0.0097 1.9064 D4ADD7 GLRX5 Glutaredoxin 5 16.40 −1.2978 0.0019 1.5977 D4A4L5 ISCA2 Iron-sulfur cluster assembly 2 16.70 −1.3870 0.0098 1.6936 G3V6P7 MYH9 Myosin, heavy polypeptide 9, nonmuscle 226.30 1.8328 0.0019 2.4356 A0A0G2KAQ5 MYOZ2 Myozenin 2 29.80 −1.4006 0.0084 1.7298 Q5XI78 OGDH 2-oxoglutarate dehydrogenase, mitochondrial 116.20 1.3004 0.0043 1.5656 P04916 RBP4 Retinol-binding protein 4 23.20 −1.3708 0.0037 1.7246 G3V8R0 RGD1311703 Similar to sid2057p 19.90 −1.5060 0.0053 1.9425 F1LSW7 RPL14 60S ribosomal protein L14 23.30 1.3440 0.0021 1.6981 Q6IRH6 SLC25A3 Phosphate carrier protein, mitochondrial 39.60 1.3677 0.0098 1.5331 P97849 SLC27A1 Long-chain fatty acid transport protein 1 71.20 1.2016 0.0037 1.5028 P07632 SOD1 Superoxide dismutase[Cu-Zn] 15.90 −1.39760 0.0040 1.7717 P62078 TIMM8B Mitochondrial import inner membrane translocase subunit Tim8 B 9.30 −1.5168 0.0073 1.9335 P62074 TIMM10 Mitochondrial import inner membrane translocase subunit Tim10 10.30 −1.2840 0.0043 1.5285 Q5XIK2 TMX2 Thioredoxin-related transmembrane protein 2 33.80 1.4949 0.0031 1.9565 B0K010 TXNDC17 Thioredoxin domain-containing 17 14.10 −1.3020 0.0052 1.5612 Q5M9I5 UQCRH Cytochrome b-c1 complex subunit 6 mitochondrial 10.40 −1.4631 0.0037 1.8937

CAFH, metabolic syndrome rats with hesperidin supplementation; CAFV, metabolic syndrome rats without hesperidin supplementation; MW, molecular weight; FC, fold change; NA, not available. a)Results from Wilcoxon tests and t-test. A p-value < 0.01 was considered statistically signiﬁcant. b)VIP score was from PLS-DA.

Network of the Kidney Tissue Proteome: The top network found downregulated proteins (in red and green, respectively) are by IPA was “Cellular compromise, free radical scavenging, cell represented in Figure 2. The proteins implicated in the top death, and survival” (score = 14), followed by “Cell death and networks after hesperidin treatment for eight weeks were survival, free radical scavenging, organismal injury, and abnor- EPB41L3, TXN2, SOD1, TPM3, NUCB2, VDAC3, MYO1D, malities” (score = 6), and “Cardiovascular system development SLC25A3, OGDH, CPT1A, MME, CDC42, ABCG2, and RACK1. and function, immunological disease, inﬂammatory disease” The other proteins implicated in the top networks related (score = 1). The overlapping graphical representation of the two to the proteins analyzed in the present study were TNF, more important networks “Cellular compromise, free rad- AGT, INSULIN, OTOF, SLC27A1, FRIN2B, TPM1, CD36, ical scavenging, cell death, and survival” and “Cardiovas- ITGB1, KCNJ11, CHMP2B, PLEC, KRT8, TPM2, MAPK14, cular system development and function, immunological HSPA5, PRKCZ, PAR6, NCF1, ERN1, EIF2AK3, CTSV, and disease, inﬂammatory disease”, as well as the up- and TP53.

Figure 1. Graphical representation of the top networks of the heart tissue proteome after hesperidin supplementation. Interaction between the differentially expressed proteins and other important proteins. Down- and upregulated proteins are symbolized in red and green, respectively. ABR, active break- point cluster region-related protein; ALDH7A1, aldehyde dehydrogenase 7 family member 1; AP2B1, AP-2 complex subunit beta; APOA4, apolipoprotein A4; ATP1A1, ATPase Na+/K+ transporting subunit alpha 1; ATP1B1, sodium/potassium-transporting ATPase subunit beta-1; ATP5F1D, ATP synthase F1 subunit delta; CACNA1B, voltage-dependent N-type calcium channel subunit alpha-1B; CAMKII, calcium/calmodulin-dependent protein kinase type II alpha chain; CDC42, cell division cycle 42; CLTC, clathrin heavy chain; CPT1A, carnitine palmitoyltransferase 1A; EPB41L2, erythrocyte membrane protein band 4.1-like 2; F11R, junctional adhesion molecule A; FILIP1, filamin-A-interacting protein 1; GRM4, metabotropic glutamate receptor 4; IL1B, interleukin-1 beta; ITSN1, intersection-1; FLNA, filamin A; KCNJ11, ATP-sensitive inward rectifier potassium channel 11; KCNJ2, inward rectifier potassium channel 2; LAMC1, laminin subunit gamma 1; MYO1C, unconventional myosin-Ic; NEFH, neurofilament heavy polypeptide; NFKB, nuclear factor NF-kappa-B p105 subunit; OGDH, oxoglutarate dehydrogenase; OTOF, otoferlin; PIN, peptidyl-propyl cis-trans isomerase NIMA-interacting 1; PP2A, serine/threonine-protein phosphatase 2A 55 kDa regulatory subunit B beta isoform; PRKCD, protein kinase C delta type; SRC, proto-oncogene tyrosine- protein kinase Src; PRKCZ, protein kinase C zeta type; RCAN1, calcipressin-1; GSK3B, glycogen synthase kinase-3 beta; SIPI, secretory leukocyte protease inhibitor; AR, androgen receptor; SLC1A2, excitatory amino acid transporter 2; SLC25A3, solute carrier family 25 member 3; SLC27A1, long-chain fatty acid transport protein 1; SOD1, superoxide dismutase 1; STAT3, signal transducer and activator of transcription 3; TNF, tumor necrosis factor; TNFRSF11B, tumor necrosis factor receptor superfamily member 11B. Figure reproduced with permission under the terms of the CC-BY 4.0 license. Copyright 2019, QIAGEN Silicon Valley.

3.3. Top Diseases and Functions Determined from the Heart and 4. Discussion Kidney Tissue Proteomes The current study presents the effects of hesperidin supplemen- −1 Table 3 details the top relevant diseases and functions involving tation of 100 mg kg body weight per day for eight weeks on the significant proteins in both heart and kidney tissues. The top the proteomic profiles of heart and kidney tissues in rats with or categories are the cardiovascular system, free radical scavenging, without MS. Proteomic analysis revealed significant changes in lipid metabolism, glucose metabolism, and renal and urological the proteomic profiles of MS rats fed a cafeteria diet with and diseases. without hesperidin supplementation in both tissues after eight weeks.

Table 2. 53 proteins diﬀerentially expressed in kidney tissue after hesperidin treatment in CAFH compared to CAFV rats.

UniProt code Gene symbol Protein description MW [kDa] Univariate analysis Multivariate analysis

FC p-valuea) VIPb)

Q80W57 ABCG2 ATP-binding cassette subfamily G member 2 72.90 1.2570 0.0031 1.6966 Q6P2A5 AK3 GTP:AMP phosphotransferase AK3, mitochondrial 25.50 −1.2561 0.0094 1.7174 Q9WUC4 ATOX1 Copper transport protein ATOX1 7.30 −1.2888 0.0021 1.8082 Q03344 ATP5IF1 ATPase inhibitor, mitochondrial 12.20 −1.3651 0.0005 2.0549 P21571 ATP5PF ATP synthase−coupling factor 6, mitochondrial 12.50 −1.6119 0.0028 2.4703 B2GUV5 ATP6V1G1 V−type proton ATPase subunit G 13.70 −1.4949 0.0014 2.2956 Q5I0M1 APO H Apolipoprotein H 38.40 −1.5042 0.0008 2.3359 F1LRS8 CD2AP CD2−associated protein 70.40 −1.2746 0.0088 1.6887 D3ZD09 COX6B1 Cytochrome c oxidase subunit 10.10 −1.4006 0.009 1.9883 P32198 CPT1A Carnitine O−palmitoyltransferase 1, liver isoform 88.10 1.4794 0.0043 2.1983 P97829 CD47 Leukocyte surface antigen CD47 33.00 1.2059 0.0038 1.5247 Q8CFN2 CDC42 Cell division control protein 42 homolog 21.20 1.2033 0.0033 1.5261 D3ZUX5 CHCHD3 MICOS complex subunit 26.40 −1.3168 0.0022 1.8771 A0A0H2UHL6 CTSH Pro−cathepsin H 32.90 −1.6982 0.0038 2.5675 P07154 CTSL Cathepsin L1 37.60 1.5305 0.0035 2.3057 Q68FR9 EEF1D Elongation factor 1−delta 31.30 −1.2527 0.0035 1.6775 A3E0T0 EPB41L3 Erythrocyte membrane protein band 4.1−like 3 96.90 −1.5757 0.0065 2.3375 P80299 EPHX2 Bifunctional epoxide hydrolase 2 62.30 4.2663 0.0087 4.1318 C0JPT7 FLNA Filamin A 280.30 1.2067 0.005 1.5157 P19468 GCLC Glutamate-cysteine ligase catalytic subunit 72.60 1.3370 0.0008 1.9709 D3ZK97 H3F3C Histone H3 15.30 1.2772 0.0097 1.691 F1M9B2 IGFBP7 Insulin-like growth factor binding protein 7, isoform CRA_b 28.90 −1.6947 0.0024 2.593 D4A4L5 ISCA2 Iron-sulfur cluster assembly 2 16.70 −1.3918 0.0055 2.0032 B2RZ79 ISCU Iron-sulfur cluster assembly enzyme 18.00 −1.3077 0.0013 1.8767 D3ZCZ9 LOC100912599 NADH dehydrogenase [ubiquinone] iron-sulfur protein 6, 13.00 −1.2978 0.0037 1.8041 mitochondrial Q562C6 LZTFL1 Leucine zipper transcription factor-like protein 1 34.60 −1.5810 0.0012 2.4546 D3Z900 MARC2 Mitochondrial amidoxime reducing component 2 38.20 1.2501 0.0009 1.7259 A0A0H2UHX5 MME Neprilysin 78.60 1.2870 0.0062 1.745 Q63357 MYO1D Unconventional myosin-Id 116.00 1.4044 0.0072 2.0134 G3V8R1 NUCB2 Nucleobindin 2. isoform CRA_b 50.10 −1.3538 0.0042 1.9359 Q5XI78 OGDH 2-oxoglutarate dehydrogenase, mitochondrial 116.20 1.2142 0.0003 1.6381 P51583 PAICS Multifunctional protein ADE2 47.10 1.4550 0.0078 2.1091 D3ZD40 PAPLN Papilin. proteoglycan-like sulfated glycoprotein 138.50 −1.4040 0.0046 2.0433 B0BN18 PFDN2 Prefoldin subunit 2 16.60 −1.2527 0.0065 1.6467 P85973 PNP Purine nucleoside phosphorylase 32.30 1.3718 0.0005 2.0718 P10960 PSAP Prosaposin 61.10 −1.4641 0.0057 2.151 P63245 RACK1 Receptor of activated protein C kinase 1 35.10 1.2130 0.0005 1.6195 Q6TXG7 SHMT1 Serine hydroxymethyltransferase 75.30 1.2142 0.0042 1.5493 F1LZW6 SLC25A13 Solute carrier family 25 member 13 54.10 1.2303 0.0023 1.6277 G3V741 SLC25A3 Phosphate carrier protein. mitochondrial 39.50 1.2050 0.0057 1.5022 Q9JJ19 SLC9A3R1 Na(+)/H(+) exchange regulatory cofactor NHE-RF1 38.80 −1.2588 0.0093 1.6398 G3V6D9 SLC9A3R2 Na(+)/H(+) exchange regulatory cofactor NHE-RF 37.30 −1.2467 0.0061 1.6295 P07632 SOD1 Superoxide dismutase[Cu-Zn] 15.90 −1.5801 0.0036 2.3891 A0A0G2K9 × 1 SPP2 Secreted phosphoprotein 24 14.90 −1.5094 0.004 2.2601 O70257 STX7 Syntaxin-7 28.80 −1.3122 0.0085 1.7901 P62078 TIMM8B Mitochondrial import inner membrane translocase 9.30 −1.5241 0.0058 2.2588 subunit Tim8 B Q63610 TPM3 Tropomyosin alpha-3 chain 29.00 −1.4241 0.0036 2.1009

(Continued)

Table 2. Continued.

UniProt code Gene symbol Protein description MW [kDa] Univariate analysis Multivariate analysis

FC p-valuea) VIPb)

P09495 TPM4 Tropomyosin alpha-4 chain 28.50 −1.3918 0.0056 2.0044 P63029 TPT1 Translationally controlled tumor protein 19.40 −1.4459 0.0027 2.163 P97615 TXN2 Thioredoxin. mitochondrial 18.20 −1.3698 0.0075 1.9353 Q5M9I5 UQCRH Cytochrome b-c1 complex subunit 6, mitochondrial 10.40 −1.4015 0.0027 2.0703 Q9Z269 VAPB Vesicle-associated membrane protein-associated protein B 26.90 −1.3168 0.0097 1.7924 A0A0G2JSR0 VDAC3 Voltage-dependent anion-selective channel protein 3 30.80 1.2226 0.0006 1.6488

CAFH, metabolic syndrome rats with hesperidin supplementation; CAFV, metabolic syndrome rats without hesperidin supplementation; MW, molecular weight; FC, fold change. a)Results from Wilcoxon tests and t-tests. A p-value < 0.01 was considered statistically signiﬁcant. b)VIP score was from PLS-DA.

In the heart tissues of MS rats, 35 proteins were differentially as atherosclerosis.[35] In the heart and kidney tissues of MS rats, expressed: 19 proteins were downregulated and 16 were upregu- another protein upregulated by hesperidin was FLNA, a large lated. Moreover, in the kidney tissues of MS rats, 53 proteins were cytoplasmic protein (1.54-fold in heart tissue and 1.20-fold in differentially expressed: 33 proteins were downregulated and 20 kidney tissue). FLNA can promote or suppress cell processes were upregulated. important for heart development.[36] FLNA is downregulated in Currently, there are no data on the effects of hesperidin rats with coronary microembolization,[37] and the lack of FLNA polyphenol on the proteomes of heart and kidney tissues in demonstrates its importance during morphogenesis of several or- healthy or MS rats. However, several studies have shown that gans, such as the heart; the lack of FLNA can result in cardiovas- polyphenols, such as resveratrol found in red wine and grapes cular malformations.[38] Moreover, the lack of editing in FLNA and secoiridiods and hydroxytyrosol present in olive oil, can increased RhoA/Rock and PLC/PKC signaling, increased aortic change the proteomes of cells, in rats and humans, improving hypercontraction and induced cardiomyocyte hypertrophy, in- different cardiovascular risk parameters such as inflammation, creasing the diastolic blood pressure.[39] cholesterol homeostasis, oxidation, and blood coagulation.[26–28] Our results suggested that hesperidin downregulated CD2AP Our results suggest that, in MS rats, hesperidin supplementa- inducing changes in the expression of proteins related to glu- tion could exert cardioprotective effects by upregulating the ex- cose metabolism. CD2AP, a cytosolic protein that interacts with pression of proteins related to the cardiovascular system such signaling molecules,[40] was downregulated in the kidney tis- as ATP1A1 (1.32-fold). ATP1A1, found in heart tissue, is asso- sues (−1.27-fold) of MS rats. Evidence suggests that high lev- ciated with ischemic acute renal failure when it is downregu- els of CD2AP increase the risk of renal disease in patients lated and with a decrease in blood pressure levels when it is with diabetes.[41] Another differentially expressed and upregu- upregulated.[29] In addition, hesperidin can be cardioprotective lated protein in kidney tissue was EPHX2 (4.26-fold). Decreased by downregulating the expression levels of proteins known to af- EPHX2 expression is related to increased insulin sensitivity in fect blood pressure, such as ATP5PF (−1.61-fold), which is found humans with MS.[42] Accordingly, hesperidin supplementation in kidney tissue and related to vasoconstriction, hypertension, could exert preventive effects on glucose metabolism in MS rats, and cardiac hypertrophy in rats and humans.[30,31] Another re- but the effects of hesperidin intake on humans with CVRFs such lated protein that appeared to be downregulated in the kidney as diabetes and MS need to be confirmed. tissues of MS rats was TPT1 (−1.44-fold). TPT is implicated in Although the present study showed that hesperidin supple- the pathogenesis of atherosclerosis and pulmonary artery hyper- mentation can change the proteomic profile to exert positive ef- tension to prevent macrophage apoptosis in the artery intima.[32] fects on different parameters in MS rats, some differentially ex- IGFBP7 was also downregulated by hesperidin in kidney tissue pressed proteins showed negative effects. For example, in MS (−1.69-fold). At high concentrations, this molecule is related to rats, the following two antioxidant proteins were downregu- poor diastolic function and ventricular systolic pressure. Thus, lated: SOD1 (−1.39-fold in heart tissue; 1.58-fold in kidney tis- elevated IGFBP7 levels could be a biomarker of diastolic dysfunc- sue) and TXN2 (−1.36-fold in kidney tissue), which likely de- tion and functional capacity in humans with heart failure.[33] creased the protection against oxidative stress in MS rats. In Moreover, our findings suggest that hesperidin could also ex- addition, SLC9A3R1 was downregulated in kidney tissue (−1.25- ert cardioprotective effects in MS rats by upregulating the expres- fold). Decreased expression of SLC9A3R1 was found in hyper- sion levels of proteins related to free radical scavenging, such as tensive rats,[43] indicating that hesperidin supplementation does ALDH7A1 (1.28-fold). ALDH7A1, found in heart tissue, protects not exert a beneficial effect on it, at least not by this pathway. cells against oxidative stress by metabolizing lipid peroxidation- However, in MS rats, some proteins related to lipid metabolism derived aldehydes produced during oxidative stress and xeno- were downregulated in the heart, such as APOA4 (−1.50-fold) biotics metabolism.[34] During lipid peroxidation, large quanti- and APOC2 (−1.83-fold). APOA4 increases triglyceride pro- ties of aldehydes are produced, and they can covalently bind to duction and reduces hepatic lipids;[44] therefore, a decrease in proteins and DNA, inactivating different proteins and damaging its expression would be expected to interfere with the cor- DNA. However, aldehydes are related to several diseases, such rect absorption and elimination of dietary fats. However, actual

Figure 2. Graphical representation of the top networks of the kidney tissue proteome after hesperidin supplementation. Interaction between the differentially expressed proteins and other important proteins. Down- and upregulated proteins are symbolized in red and green, respectively. ABCG2, ATP binding cassette subfamily G member 2; AGT, angiotensinogen; CD36, CD36 molecule (trombospondin receptor); CDC42, cell division cycle 42; CHMP2B, charged multivesicular body protein 2B; CPT1A, carnitine palmitoyltransferase 1A; CTSV, cathepsin V; EIF2AK3, eukaryotic translation initi- ation factor 2-alpha kinase 3; EPB41IL3, erythrocyte membrane protein band 4.1-like 3; ERN1, endoplasmic reticulum to nucleus-signaling 1; GRIN2B, glutamate receptor ionotropic NMDA 2B; HSPA5, endoplasmic reticulum chaperone BiP; ITGB1, integrin beta-1; KCNJ11, ATP-sensitive inward rectifier potassium channel 11; KRT8, keratin type II cytoskeletal 8; MAPK14, mitogen-activated protein kinase 14; MME, neprilysin; MYO1D, unconventional myosin-Id; NCF1, neutrophil cytosolic factor 1; NUCB2, nucleobindin 2; OGDH, oxoglutarate dehydrogenase; OTOF, otoferlin; PAR6, partitioning de- fective 6 homolog alpha; PLEC, plectin; PRKCZ, protein kinase C zeta type; RACK1, receptor of activated protein C kinase 1; SLC25A3, solute carrier family 25 member 3; SLC27A1, solute carrier family 27 member 1; SOD1, superoxide dismutase 1; TNF, tumor necrosis factor; TP53, cellular tumor antigen o53; TPM1, tropomyosin alpha-1 chain; TPM2, tropomyosin beta chain; TPM3, tropomyosin alpha-3 chain; TXN2, thioredoxin mitochondrial; VDAC3, voltage-dependent anion-selective channel protein 3. Figure reproduced with permission under the terms of the CC-BY 4.0 license. Copyright 2019, QIAGEN Silicon Valley. evidence has demonstrated that decreased levels of APOA4 in- Additionally, in MS rats after administered hesperidin sup- crease the chylomicron size, delaying its clearance from the blood plementation, some of the proteins differentially expressed in and indicating it is not required for triglyceride absorption in the heart and kidney tissues were related to cancer. In the heart the mouse intestine.[45] Thus, more studies of hesperidin sup- tissues of MS rats, CCDC58 and SLC25A3 were upregulated plementation are needed to clarify these effects on the proteomic (1.73-fold and 1.36-fold, respectively). CCDC58 is a biomarker profile. of breast, endometrial, and urethral cancer, and SLC25A3 is

Table 3. Top diseases and functions determined by heart and kidney tissue proteomic analyses in CAFH compared to CAFV rats.

Categories Heart tissue proteins Kidney tissue proteins

Cardiovascular system Development of neointima – ↓TPT1 Blood pressure – ↓ATP5PF, ↑EPHX2, ↓SOD1 Density of microvessel – ↑EPHX2 Angiogenesis – ↑Cdc42, ↑EPHX2, ↓TPT1 Binding of endothelial cells – ↑CD47 Sprouting angiogenesis ↑Cdc42 ↑CDC42 Heart rate ↑ATP1A1, ↓SOD1 – Ischemic acute renal failure ↑ATP1A1 – Free radical scavenging Metabolism, removal, and quantity of ↓SOD1 ↓SOD1 superoxide Biosynthesis of hydrogen peroxide ↓SOD1 – Lipid metabolism Fatty acid metabolism ↑OGDH, ↓APOA4, ↑APOC1, ↑GCDH, ↑OGDH, ↓APOH, ↓ATP5PF, ↑CPT1A, ↓RBP4, ↑SLC27A1 ↑EPHX2, ↓IGFBP7 Transport of lipid ↓APOA4, ↑APOC1, ↓RBP4, ↑SLC27A1 – Metabolism of lipoprotein ↑APOC1 – Synthesis of epoprostenol – ↓ATP5PF, ↓IGFBP7 Synthesis of prostaglandin – ↓ATP5PF, ↑EPHX2, ↓IGFBP7 Binding of eicosapentenoic acid and – ↑CPT1A malonyl-coenzyme A Beta-oxidation of oleic acid – ↑CPT1A Transport of triacylglycerol – ↓APOH Quantity of long-chain acyl-coenzyme A – ↑CPT1A Metabolism of succinyl-coenzyme A ↑OGDH ↑OGDH Accumulation of triacylglycerol – ↑CPT1A Oxidation of fatty acid ↑GCDH, ↑SLC27A1 ↑CPT1A, ↓NUCB2 Metabolism of long chain fatty acids – ↑CPT1A Esteriﬁcation, transport and oxidation of ↑SLC27A1 ↑CPT1A palmitic acid Synthesis and metabolism of acyl-coenzyme A ↑GCDH, ↑OGDH – Transport of retinol ↓RBP4 – Glucose metabolism Insulin sensitivity index – ↓NUCB2 Secretion of glucagon – ↓NUCB2 Renal and urological disease Nephrosis ↑CLTC – Apoptosis of kidney cells ↑ATP1A1, ↑CDC42 –

CAFH, metabolic syndrome rats with hesperidin supplementation; CAFV, metabolic syndrome rats without hesperidin supplementation; —, no protein identified; ↑, protein upregulated; ↓, protein downregulated. overexpressed in cervical carcinomas.[46] In the kidney tissues The present study has several limitations that warrant discus- of MS rats, the proteins UQCRH and LZTFL1 were downregu- sion. One limitation is the small sample size. Based on the results lated (−1.40-fold and −1.58-fold, respectively). In the literature, obtained in the present work, further studies with larger sample UQCRH overexpression is associated with a poor prognosis for sizes and other experimental models, such as diabetes and can- lung cancer and hepatocellular carcinoma patients.[47,48] Finally, cer, will be performed to confirm the observed effects. Addition- LZTFL1 is a tumor suppressor and an independent prognos- ally, the dose of hesperidin should be increased or decreased to tic marker for the survival of gastric cancer patients when it observe the different effects of various doses, which will be based is elevated.[49] In this sense, several studies in cancer models on the natural doses of hesperidin present in citrus fruits or juice have shown that hesperidin can delay cell proliferation,[50] in- to extrapolate the results from rats to humans. hibit cell viability, and induce apoptosis in cancer cells.[51] There- In conclusion, hesperidin supplementation for eight weeks fore, studies of hesperidin supplementation in rats with cancer can change the proteomic profiles of the heart and kidney tissues and its effects on the proteome are needed to provide more ro- in MS rats and has a beneficial impact on the cardiovascular sys- bust evidence and to clarify whether hesperidin could be a tumor tem, free radical scavenging, and lipid and glucose metabolism. suppressor. Therefore, the identification of proteins involved in metabolic

Mol. Nutr. Food Res. 2020, 64, 1901063 1901063 (10 of 12) © 2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.advancedsciencenews.com www.mnf-journal.com pathways can help to understand the molecular basis of hes- [11] C. Wunpathe, P. Potue, P. Maneesai, S. Bunbupha, P. Prachaney, U. peridin in MS rats. However, further research is needed to con- Kugongviriyapan, V. Kugongviriyapan, P. Pakdeechote, Am. J. Chin. firm the results reported in the present study in humans. Med. 2018, 46, 751. [12] L. Dobiaš, M. Petrová, R. Vojtko, V. Kristová, Phytother. Res. 2016, 30, 1665. [13] S. Jia, Y. Hu, W. Zhang, X. Zhao, Y. Chen, C. Sun, X. Li, K. Chen, Food Supporting Information Funct. 2015, 6, 878. [14] Y. Z. Sun, J. F. Chen, L. M. Shen, J. Zhou, Y. Chen, C. Sun, X. Li, K. Supporting Information is available from the Wiley Online Library or from Chen, 2017, , 109. the author. Eur. J. Pharmacol. 815 [15] H. Iskender, E. Dokumacioglu, T. M. Sen, I. Ince, Y. Kanbay, S. Saral, Biomed. Pharmacother. 2017, 90, 500. [16] P. Maneesai, S. Bunbupha, P. Potue, T. Berkban, U. Kugongviriyapan, Acknowledgements V. Kukongviriyapan, P. Prachaney, P. Pakdeechote, Nutrients 2018, 10, 1. L.A., R.S., and A.P. contributed equally to this work. This research was [17] M. J. Lerma-García, A. D’Amato, E. F. Simó-Alfonso, P. G. Righetti, E. co-funded by the 7th Framework Program of the European Union and Fasoli, Food Chem. 2016, 196, 739. the Agency for Business Competitiveness of the Government of Catalonia [18] J.-H. Wang, J.-J. Liu, K.-L. Chen, H.-W. Li, J. He, B. Guan, L. He, BMC (ACCIÓ) under the TECNOspring program awarded to JM-P (TECSPR14- Genomics 2017, 18, 984. 2-0001). A.P. has a Torres Quevedo contract (Subprograma Estatal de Incorporación, Plan Estatal de Investigación Científica y Técnica y de In- [19]D.F.S.Chaves,P.C.Carvalho,E.Brasili,M.M.Rogero,N.M.A. novación). U.C. has a Pla Estratègic de Recerca i Innovació en Salut post- Hassimotto, J. K. Diedrich, J. J. Moresco, J. R. Yates, F. M. Lajolo, J. doctoral grant (SLT0021/16/00239); Catalunya, Spain). Proteome Res. 2017, 16, 4086. [20] S. Yumnam, V. V. G. Saralamma, S. Raha, H. J. Lee, W. Suplee, E. H. Kim, S. J. Lee, J. D. Heo, G. S. Kim, Mol. Med. Rep. 2017, 16, 5386. [21] M. Guirro, A. Costa, A. Gual-Grau, J. Mayneris-Perxachs, H. Torrell, Conflict of Interest P. Herrero, N. Canela, L. Arola, Electrophoresis 2018, 39, 1692. [22] A. Mosqueda-Solís, J. Sánchez, B. Reynés, M. Palou, M. P. Portillo, A. The authors declare no conflict of interest. Palou, C. Picó, Sci. Rep. 2018, 8, 15100. [23] S. Reagan-Shaw, M. Nihal, N. Ahmad, FASEB J. 2008, 22,659. [24] M. Guirro, A. Gual-grau, A. Gibert-ramos, J. M. Alcaide-hidalgo, N. Author Contributions Canela, L. Arola, J. Mayneris-Perxachs, Antioxidants 2020, 9, 79. [25] K. G. M. M. Alberti, R. H. Eckel, S. M. Grundy, P. Z. Zimmet, J. I. M.G., A.G.-G., A.G.R., and J.M.-P. performed the experimental study in Cleeman, K. A. Donato, J. C. Fruchart, W. P. T. James, C. M. Loria, S. rats. N.C. and E.F.-R. performed the proteomic analysis. L.P.-P. analyzed C. Smith, Circulation 2009, 120, 1640. the proteomic results under the supervision of U.C. and L.P.-P. wrote the [26] M. K. Nøhr, T. P. Kroager, K. W. Sanggaard, A. D. Knudsen, A. Stens- manuscript. U.C., N.C., J.M.-P., R.M.V., L.L.A., R.S., and A.P. provided feed- balle, J. Enghild, J. Olholm, B. Richelsen, S. B. Pedersen, PLoS One back on the manuscript. 2016, 11, e0159747. [27] Ú. Catalán, L. Rubió, M. C. L. de Las Hazas, P. Herrero, P. Nadal, N. Canela, A. Pedret, M. J. Motilva, R. Solà, Mol. Nutr. Food Res. 2016, Keywords 60, 2114. [28] A. Pedret, Ú. Catalán, S. Fernández-Castillejo, M. Farràs, R. M. Valls, hesperidin, kidneys, metabolic syndrome, proteomics L. Rubió, N. Canela, G. Aragonés, M. Romeu, O. Castañer, R. Torre, M. I. Covas, M. Fitó, M. J Motilva, R. Solà, PLoS One 2015, 10, e0129160. Received: October 10, 2019 [29] T. J. Pritchard, M. Parvatiyar, D. P. Bullard, R. M. Lynch, J. N. Lorenz, Revised: January 29, 2020 R. J. Paul, Am. J. Physiol.: Heart Circ. Physiol. 2007, 293, H1172. Published online: May 4, 2020 [30] T. Osanai, S. Sasaki, T. Kamada, N. Fujiwara, T. Nakano, H. Tomita, T. Matsunaga, K. Magota, K. Okumura, J. Hypertens. 2003, 21, 2323. [31] T. He, A. Guan, Y. Shi, Z. Ge, H. Dai, Herz 2015, 40, 783. [1] C.N.Zhao,X.Meng,Y.Li,S.Li,Q.Liu,G.Y.Tang,H.B.Li,Nutrients [32] D. Pinkaew, K. Fujise, Adv. Clin. Chem. 2017, 82, 265. 2017, 9, E598. [33] P. U. Gandhi, H. K. Gaggin, M. M. Redfield, H. H. Chen, S. R. Stevens, [2] E.M.Alissa,G.A.Ferns,Crit. Rev. Food Sci. Nutr. 2017, 57, 1950. K. J. Anstrom, M. J. Semigran, P. Liu, J. L. Januzzi Jr., JACC Heart Fail. [3] Top 10 Causes of Death, World Health Organization, May 2018. 2016, 4, 860. [4] A. Serino, G. Salazar, Nutrients 2019, 11, E53. [34] C. Brocker, M. Cantore, P. Failli, V. Vasiliou, Chem.-Biol. Interact. 2011, [5] J. Rienks, J. Barbaresko, U. Nöthlings, Nutrients 2017, 9, 415. 191, 269. [6] Z. Cui, S. Dewey, A. V. Gomes, Am. J. Cardiovasc. Dis. 2011, 1, 274. [35] A. Negre-Salvayre, N. Auge, V. Ayala, H. Basaga, J. Boada, R. Brenke, [7] M. Mokou, V. Lygirou, A. Vlahou, H. Mischak, Expert Rev. Proteomics S. Chapple, G. Cohen, J. Feher, T. Grune, G. Lengyel, G. E. Mann, 2017, 14, 117. R. Pamplona, G. Poli, M. Portero-otin, Y. Riahi, R. Salvayre, S. Sas- [8] M. Yamamoto, H. Jokura, K. Hashizume, H. Ominami, Y. Shibuya, A. son, J. Serrano, O. Shamni, W. Siems, R. C. M. Siow, I. Wiswedel, K. Suzuki, T. Hase, A. Shimotoyodome, Food Funct. 2013, 4, 1346. Zarkovic, N. Zarkovic, Free Radical Res. 2010, 44, 1125. [9] F. Homayouni, F. Haidari, M. Hedayati, M. Zakerkish, K. Ahmadi, Phy- [36] Etiology and Morphogenesis of Congenital Heart Disease (Eds: T.Nakan- tother. Res. 2018, 32, 1073. ishi, R. R. Markwald, H. S. Baldwin, B. B. Keller, D. Srivastava, H. [10] O. D. Rangel-Huerta, C. M. Aguilera, M. V. Martin, M. J. Soto, M. C. Yamagishi), Springer, Japan 2016. Rico, F. Vallejo, F. Tomas-Barberan, A. J. Perez-de-la-Cruz, A. Gil, M. [37] A. Chen, Z. Chen, Y. Xia, D. Lu, J. Jia, K. Hu, A. Sun, Y. Zou, J. Qian, J. D. Mesa, J. Nutr. 2015, 145, 1808. Ge, Front. Physiol. 2018, 9, 1318.

[38] K. Toomer, K. Sauls, D. Fulmer, L. Guo, K. Moore, J. Glover, R. [43] K. Kobayashi, T. Monkawa, M. Hayashi, T. Saruta, J. Hypertens. 2004, Stairley, J. Bischoﬀ, R. A. Levine, R. A. Norris, Anat. Rec. 2019, 302, 22, 1723. 117. [44] M. A. Verhague, D. Cheng, R. B. Weinberg, G. S. Shelness, Arte- [39] M. Jain, T. D. Mann, M. Stuli´c, S. P. Rao, A. Kirsch, D. Pullirsch, S. rioscler., Thromb., Vasc. Biol. 2013, 33, 2501. Strobl, C. Rath, L. Reissing, K. Moreth, T. Klein-Rodewald, R. Bekered- [45] A. B. Kohan, F. Wang, X. Li, A. E. Vandersall, S. Huesman, M. Xu, Q. jian, V. Gailus-Durner, H. Fuchs, M. H. Angelis, E. Pablik, L. Cimatti, Yang, D. Lou, P. Tso, Am. J. Physiol. Gastrointest. Liver Physiol. 2013, D. Martin, J. Zinnanti, W. F. Graier, M. Sibilia, S. Frank, E. Y. Levanon, 304, G1128. M. F. Jantsch, EMBO J. 2018, 37, e94813. [46] F. Palmieri, Mol. Aspects Med. 2013, 34, 465. [40] T. B. Huber, B. Hartleben, J. Kim, M. Schmidts, B. Schermer, A. Keil, [47] F. Gao, Q. Liu, G. Li, F. Dong, M. Qiu, X. Lv, S. Zhang, Z. Guo, Open L. Egger, R. L. Lecha, C. Borner, H. Pavenstadt, A. S. Shaw, G. Walz, Biol. 2016, 6, 150256. T. Benzing, Mol. Cell. Biol. 2003, 23, 4917. [48] E. R. Park, S. B. Kim, J. S. Lee, Y. H. Kim, D. H. Lee, E. H. Cho, S. H. [41] M. E. Hyvönen, P. Ihalmo, N. Sandholm, M. Stavarachi, C. Fars- Park, C. J. Han, B. Y. Kim, D. W. Choi, Y. D. Yoo, A. Yu, J. W. Lee, J. J. blom, A. J. McKnight, M. Lajer, A. Maestroni, G. Lewis, L. Tanow, S. Jang, Y. N. Park, K. S. Suh, K. H. Lee, Cancer Med. 2017, 6, 749. Maestroni, G. Zerbini, H. H. Parving, A. P. Maxwell, P. H. Groop, S. [49] Q. Wei, W. Zhou, W. Wang, B. Gao, L. Wang, J. Cao, Z. P. Liu, Cancer Lehtonnen, Acta Diabetol. 2013, 50, 887. Res. 2010, 70, 2942. [42] C. E. Ramirez, M. M. Shuey, G. L. Milne, K. Gilbert, N. Hui, C. Yu, [50] J. M. P. Ferreira de Oliveira, C. Santos, E. Fernandes, Phytomedicine J. M. Luther, N. J. Brown, Prostaglandins Other Lipid Mediators 2014, 2019, 10, 152887. 113–115, 38. [51] J. Zhao, Y. Li, J., Gao, Y. De, Oncol. Lett. 2017, 14, 5569.

5. SUMMARY OF RESULTS

280

281

5. 1 Project 1. Effects of hesperidin consumption on cardiovascular risk biomarkers: a systematic review of animal studies and human randomized clinical trials.

From 698 articles, after the study selection according to our eligibility criteria and search strategy, twelve animal studies conducted in rats and mice published between January 2003 and January 2018 were included in the systematic review. All the animals (N= from 4 to 16) had at least one CVDRF (hypertension, type 2 diabetes, systemic inflammation, myocardial ischaemia, hypercholesterolaemia) and they consumed a daily dose of hesperidin of 5-200 mg/kg/body weight from 7 days to 24 weeks. Further details of each study are presented in Table 3 of the published version.

The results for anthropometry parameters showed that 6 studies evaluated these parameters, and only 1 of them showed a significant decrease in body weight after 4.6% of the total calories in the diet was composed of hesperidin.

The results for inflammation and oxidation biomarkers showed that 2 studies evaluated these parameters, and they showed a significant decrease in interleukin-6 (IL-6) levels after treatment with 100 mg/kg body weight hesperidin for 4 weeks in rats with in systemic inflammation, and showed significant decreases in NO levels after treatment with 50 mg/kg body weight hesperidin for 30 days in type 2 diabetic rats.

The results for glucose and insulin levels showed that 7 studies evaluated these parameters and 6 of them showed significantly lowered

282 blood glucose levels in type 2 diabetic rats and mice after 50-200 mg/kg body weight hesperidin for 15-24 days. On the other hand, no significant changes were observed in insulin levels after hesperidin consumption.

The results for the lipid profile showed that 4 studies evaluated these parameters and 3 and 2 studies showed significant decreases in TC levels and LDL-c levels, respectively, after 50-200 mg/kg body weight of hesperidin. Moreover, no significant changes were observed in HDL- c levels after hesperidin consumption.

From 1917 articles, after the study selection according to our eligibility criteria and search strategy, eleven human randomized controlled clinical trials with nutritional intervention published between January 2003 and January 2018 were included in the systematic review. All the subjects (N= from 22 to 194) had at least one CVDRF (overweight, obesity, metabolic syndrome, hypercholesterolaemia) and they consumed a dose of hesperidin of 50-200 mg/kg body weight for 1.5- 13 weeks. Further details of each study are presented in Table 4 and Table 5 of the published version (page 103 of the present doctoral thesis).

The results for anthropometric parameters showed that 3 studies evaluated the effects of hesperidin consumption on body weight, and 2 of them reported significant decreases after 54.60-582.50 mg/day of hesperidin in OJ for 12-13 weeks. Moreover, 5 studies evaluated the effects of hesperidin consumption on BMI and 2 of them reported significant decreases after 54.60-582.50 mg/day of hesperidin in OJ for

283

12-13 weeks in hypercholesterolaemic and overweight or obese subjects.

Finally, 1 study evaluated the effect of hesperidin consumption on body fat and significant decreases were observed after 13 weeks of 54.60 mg/day hesperidin in OJ in hypercholesteremic subjects.

The results for vascular parameters showed that 5 studies evaluated systolic and diastolic BP levels, and 2 of them reported significant decreases after 12 weeks of 237-292 mg/day of hesperidin consumption in OJ but also in the form of pure hesperidin capsules in overweight or obese subjects. On the other hand, 5 studies evaluated the effects of hesperidin consumption on endothelial function and 2 of them reported significant increases after 1.5-3 weeks of 159.60-500 mg/day hesperidin in OJ in subjects with metabolic syndrome.

The results for glucose and insulin levels showed that 5 and 4 studies evaluated plasma glucose levels and plasma insulin levels, respectively, and no significant changes were reported.

The results for lipid profile parameters showed that 8 studies evaluated the effect of hesperidin consumption on TC levels and LDL-c and 2 of them reported significant decreases in both parameters after 42-64.50 mg/day hesperidin or hesperetin in OJ for 8-13 weeks in overweight and hypercholesterolaemic subjects. Moreover, the effect of hesperidin consumption on HDL-c levels was evaluated in 8 studies and 1 of them reported a significant increase after 54.60 mg/day of hesperetin in OJ for 13 weeks in overweight subjects. Furthermore, 8 studies evaluated the effects of hesperidin consumption on TG levels and 1 of them reported a significant decrease after 237 mg/day of hesperidin in orange

284 juice for 12 weeks in overweight or obese subjects. Finally, the effect of hesperidin consumption on apolipoprotein A-1 and B was evaluated in 3 studies and 1 of them observed a significant increase after 213 mg/day of hesperidin in OJ for 4 weeks in hypercholesterolaemic subjects; while 1 of them reported a significant decrease after 237 mg/day of hesperidin in OJ for 12 weeks in overweight or obese subjects.

The results for coagulation biomarkers showed that 2 studies evaluated the plasma levels of fibrinogen and homocysteine after hesperidin consumption and no significant changes were observed.

The results for inflammation biomarkers showed that 1 study evaluated the effect of hesperidin consumption on plasma protein serum amyloid A, sVCAM-1, sICAM-1, sE-selectin and sP-selectin, and no significant changes were observed. On the other hand, 1 study evaluated the effect of hesperidin consumption on plasma IL-6 levels and after 159.50 mg/day of hesperidin in OJ for 1.5 weeks, the levels significantly decreased in subjects with metabolic syndrome.

Finally, the results for oxidative biomarkers showed that 2 studies evaluated NO levels, and no significant changes were observed.

285

5. 2 Project 2. Effects of hesperidin in orange juice in blood pressure and pulse pressure in mildly hypertensive individuals: a randomized clinical trial (CITRUS study).

Participants and intervention compliance: A total of 159 adult participants with pre- and stage 1 hypertension completed the 12 weeks of the randomized clinical trial: N=43 in the control group, N=46 in the OJ group and N=40 in the EOJ group; 52 participants completed the single dose study: N= 17 in CD, N=21 in OJ and N=14 in EOJ. Further details with the flow diagram are presented in Figure 1 of the published version.

Regarding the baseline characteristics, no differences were observed among the groups or in the level of physical activity at the end of the study. Regarding dietary intake, only an increased intake of protein was observed in the OJ group compared with the EOJ group. All the subjects complied with the nutritional intervention since 6 compliance biomarkers namely hesperetin-7-O-β-d-glucuronide, hesperetin-3-O-β-d-glucuronide, hesperetin-7-O-sulfate, naringerine- 4-O-β-d-glucuronide, naringenin-glucuronide and naringenin-sulfate, were significantly increased with OJ and EOJ consumption compared with the control group.

Changes in blood pressure, the main outcome: Hesperidin consumption in the OJ and EOJ groups significantly decreased the SBP levels compared to the control group. The average decreases in all SBP levels during the 12 weeks of hesperidin consumption in OJ and EOJ were -6.35- and 7.36-mm Hg, respectively.

286

Additionally, PP decreased in a dose-dependent manner with the hesperidin dose of the drinks consumed.

On the other hand, after the first single dose study significant decreases were observed in SBP levels at 2 hours after EOJ, and no effects were observed for DBP levels. After the second single dose study realized after 12 weeks of nutritional intervention, EOJ showed decreases in BP and PP levels.

Changes in secondary outcomes: homocysteine, uric acid, ICAM-1 and VCAM levels: After 12 weeks of OJ and EOJ consumption, the levels of homocysteine decreased significantly compared to CD. After the second single dose study, the homocysteine levels also decreased after 2 hours and 4 hours of OJ consumption and after 2 hours of EOJ consumption. Additionally, after 12 weeks of EOJ consumption, uric acid levels significantly decreased compared to CD. Moreover, after 12 weeks of EOJ the levels of ICAM-1, VCAM and F2-isoprostanes significantly decreased. All these changes were observed without changes in body weight and blood glucose levels.

Transcriptomic analysis: After 12 weeks of EOJ the expression of the genes PTX3 and NAMPT in PBMCs was significantly decreased compared to CD. The decreases in SBP and PP were directly correlated with the decreased expression of both genes.

287

5. 3 Project 3. Effect of hesperidin consumption in orange juice on the transcriptomic profile of subjects with pre- and stage 1 hypertension: a randomized controlled trial (CITRUS study).

Baseline characteristics of the participants: Of the 159 subjects included in the original clinical trial, 37 subjects had PBMCs collected and completed both single dose and sustained studies, making them candidates for transcriptomic profile evaluation. The 37 subjects were 41-65 years old and had pre- or stage 1 hypertension. They were from the 3 intervention groups: N=11 in the CD, N=15 in the OJ group and N=11 in the EOJ group.

Gene expression profile after single dose intervention: After 6 hours of OJ consumption in the single dose study, 3 genes were significantly different (P < 0.05) compared to CD. Of these, 2 genes were upregulated (DHRS9, related with an increase of insulin resistance; and PKDL1, with unknown function) and 1 gene was downregulated (TNFAIP3, an anti-inflammatory gene). Moreover, several genes were borderline (P < 0.10) differentially expressed (such as NFKBIA, CCL3 and CCL4L2) after OJ and EOJ consumption related to the inflammation pathways.

Gene expression profile after sustained intervention: After 12 weeks of hesperidin consumption in OJ, 12 genes were significantly different compared to CD. From these, 7 genes were upregulated: CCL20, FAM53B, LINC 01220, LncRNA SNRPD3-2, LncRNA NFKBID-1, LncRNA PDE3B-1, LOC101929524; and 5

288 genes downregulated: TNF, TMPO-AS1, BPIFB3, LncRNA ACOT-13 and CCT8-1. Moreover, 13 genes were borderline differentially expressed after OJ consumption (such as NFKBIA and IL1B). After 12 weeks of hesperidin consumption in EOJ, 18 genes were significantly different compared to CD. From these, 4 genes were upregulated: DSP, FAM53B, LncRNA SNRPD3-2 and LncRNA SLC39A8-1; and 14 genes downregulated: TNF, IL1B, CCL3L3, CXCL2, CXCL8, PTGS2, IER3, PTX3, KMT22-AS1, ATP2B1-AS1, LncRNA CCT8-1, LncRNA GRK3-1, LncRNA CCDC117-1 and LOC644090. Moreover, 17 genes were borderline differentially expressed after OJ consumption (such as ICAM, CCL20 and CXCL3).

The significant changes observed in the genes after OJ and EOJ consumption were related to an improvement in the following top diseases and functions: inflammation, cardiovascular system, acute coronary syndrome, obesity, haemoglobin, neuronal differentiation and cancer.

289

5. 4 Project 4. Hesperidin consumption in orange juice modulates plasma and urine metabolic profiles in pre- and stage 1 hypertensive subjects promoting beneficial effects on cardiovascular system: targeted and nontargeted metabolomic approach (CITRUS study).

Volunteers: A total of 159 subjects with pre- and stage 1 hypertension completed the 12 weeks of the CITRUS randomized clinical trial: N=43 in the CD, N=46 in the OJ group and N=40 in the EOJ group; 52 participants completed the single dose study: N= 17 in CD, N=21 in OJ and N=14 in EOJ. However, the targeted metabolomics analysis was performed in plasma and urine samples of 129 subjects who completed the sustained study. Nontargeted metabolomics was performed in serum samples of 52 subjects who completed both the single dose and sustained studies and in urine samples of 129 subjects.

Results of targeted metabolomics in plasma and urine samples: After 12 weeks of OJ and EOJ consumption, the plasma and urine metabolites hesperetin 7-O-B-D-glucuronide, hesperetin 3-O-B-D- glucuronide, hesperetin 7-O-sulfate, naringenin 4-O-B-D-glucuronide, naringenin glucuronide and naringenin sulfate were statistical significantly increased. Moreover, plasma and urine hesperetin 7-O-B- D-glucuronide was the main differentially expressed metabolite between both OJ and EOJ interventions. Additionally, urine hesperetin 7-O-B-D-glucuronide metabolite was inversely correlated with SBP level.

290

Nontargeted metabolomics in serum: Single dose study: After 6 hours of OJ and EOJ consumption, increased levels of proline betaine and dimethylglycine, and decreased levels of leucine were observed. Moreover, after single-dose EOJ consumption, decreased levels of urine isoleucine were observed. On the other hand, after an analysis to classify the subjects into low- and high-flavanones absorbers, increased serum levels of proline betaine, 3- hydroxybutyrate, DMG, acetoacetate and glutamine and decreased serum levels of leucine, isoleucine, valine, lysine, and alanine were observed in high flavanone absorbers.

Chronic study: After the analysis performed in a subsample of 52 subjects, who completed both the single dose study and the sustained study, increased serum levels of proline betaine and decreased serum levels of glycerophosphocholine (GPC), N-acetylglycoproteins (NAG), acetate, valine, isoleucine, and leucine were observed after OJ and EOJ consumption compared to CD.

Nontargeted metabolomics in urine: After the analysis performed in a sample of 129 subjects, the consumption of OJ for 12 weeks increased urine levels of proline betaine and decreased levels of hydroxyhippurate, pseudouridine, PAG, 4-cresyl sulfate, creatinine, DMA, NAG, alanine and 3-methyl-2- oxovalerate were observed compared to CD. After 12 weeks of EOJ consumption increased urine levels of proline betaine and decreased levels of 4-cresyl sulfate, pseudouridine, uracil, creatinine, creatine, NAG, alanine, 2-HIB and 3-MOV were observed.

291

5. 5 Project 5. Proteomic analysis of heart and kidney tissues in healthy and metabolic syndrome rats after hesperidin supplementation.

The feeding of a cafeteria diet to Sprague Dawley rats produces an obesogenic dietary pattern and increases body weight, fat mass, SBP, TG, glucose, and LDL-c, producing metabolic syndrome.

After proteomic analysis of heart tissue samples, 1127 proteins were identified. After filtering, 872 proteins were candidates for the statistical analysis. After the analysis, in healthy rats fed a standard diet no significant differences were observed between the treatment groups (between the group that received hesperidin by the vehicle and the group that did not receive hesperidin). On the other hand, in metabolic syndrome rats (the rats fed with cafeteria diet) 35 proteins were differentially expressed between the treatment groups (p<0.01 and VIP score from PLS-DA >1.5). After hesperidin consumption in metabolic syndrome rats, 19 proteins were downregulated in heart tissue samples: APOC2, COQ8A, GLRX5, ENSA, RBP4, UQCRH, SOD1, CDV3, TIMM10, ATP5F1D, TXNDC17, RGD1311703, CFL2, COX5A, TIMM8B, MYOZ2, APOA4, GCSH, ISCA2; and 16 proteins were upregulated: ALDH7A1, FLNA, MYH9, ATP1A1, RP114, CLTC, CDC42, TMX2, APOC1, SLC27A1, OGDH, ACAD8, GCDH, AFG312, CDC58 and SLC25A3.

The most important and significantly affected pathways for hesperidin supplementation in heart tissue of metabolic syndrome rats, analysed by Ingenuity Pathway Analysis (IPA), were the following: - Production of NO and reactive oxygen species in macrophages.

292

- Clathrin-mediated endocytosis signaling. - LXR/RXR activation. - Atherosclerosis signaling. - FXR/RXR activation.

On the other hand, the top networks founded by IPA were the following: - Molecular transport, carbohydrate metabolism and small molecule biochemistry. - Cardiac arrhythmia, CVD and metabolic disease. - Cancer, cardiovascular and cell cycle.

After proteomic analysis of kidney tissue samples, 1753 proteins were identified. After filtering, 1341 proteins were candidates for the statistical analysis. After the analysis, in healthy rats fed a standard diet no significant differences were observed between the treatment groups. On the other hand, in metabolic syndrome rats, 53 proteins were differentially expressed between the treatment groups (p<0.01 and VIP score from PLS-DA >1.5). After hesperidin consumption in metabolic syndrome rats, 33 proteins were downregulated in kidney tissue samples: ATP5IF1, APOH, LZTFL1, ISCU, ATP6V1G1, ATOX1, CHCHD3, IGFBP7, TPT1, UQCRH, ATP5PF, EF1D, SOD1, TPM3, LOC100912599, CTSH, SPP2, NUCB2, PAPLN, ISCA2, TPM4, PSAP, TIMM8B, SLC9A3R2, EPB4113, PFDN2, TXN2, STX7, CD2AP, COX6B1, SLC9A3R1, AK3, VAPB; and 20 proteins were upregulated: OGDH, RACK1, PNP, VDAC3, GCLC, MARC2, SLC25A13, ABCG2, CDC42, CTSL, CD47, SHTM1, CPT1A, FLNA, SLC25A3, MME, MYOLD, PAICS, EPHX2 and H3F3C. The detailed information of each protein is presented in Table 2 of the published version.

293

The most important and significantly affected pathways for hesperidin supplementation in kidney tissue of metabolic syndrome rats, analysed by IPA, were the following: - Mitochondrial dysfunction. - Xanthine and xanthosine salvage. - Guanine and guanosine salvage I. - Adenine and adenosine salvage I.

On the other hand, the top networks founded by IPA were the following: - Cellular compromise, free radical scavenging, cell death and survival. - Cell death and survival, free radical scavenging, organismal injury and abnormalities. - Cardiovascular system development and function, immunological disease, inflammatory disease.

Finally, the significant changes observed in the expressed proteins in both heart and kidney tissues of metabolic syndrome rats after hesperidin consumption were related with to an improvement in these top diseases and functions: - Cardiovascular system. - Free radical scavenging. - Lipid metabolism. - Glucose metabolism. - Renal and urological diseases.

294

6. DISCUSSION

295

296

The present work provides evidence that hesperidin consumption has beneficial effects reducing BP in humans, whereas genes expression and metabolites changes suggests effects also on others CVD risk biomarkers having protective capacity on cardiovascular health.

In this sense, the CITRUS randomized controlled trial showed the capacity of hesperidin to decrease SBP and PP, in a dose-dependent manner, in subjects with pre- and stage 1 hypertension after 12 weeks of OJ (392 mg/day of hesperidin) and EOJ (670 mg/day of hesperidin) consumption. Additionally, after single dose studies hesperidin decreased SBP and PP levels. On the other hand, markers of oxidation and inflammation (homocysteine, uric acid, ICAM-1 and VCAM) were reduced after 12 weeks of hesperidin consumption, and uric acid concentrations were directly related to SBP, DBP and PP at week 12. Moreover, daily consumption for 12 weeks and single dose consumption of hesperidin in OJ and EOJ significantly changed the transcriptome profile of PBMCs of subjects with pre- and stage 1 hypertension, with hesperidin being an anti-inflammatory agent because its capacity to downregulate pro-inflammatory genes and decrease the insulin resistance in higher doses. Furthermore, daily consumption for 12 weeks of hesperidin in OJ and EOJ changed the metabolome profile of urine and plasma/serum samples in subjects with pre- and stage 1 hypertension. Hesperidin changes endogenous metabolites related to BP, oxidative stress, inflammation and uraemic toxins, indicating anti-inflammatory and antioxidant actions, and lower SBP levels and ureamic toxins, providing beneficial effects on the cardiovascular system. Finally, daily hesperidin consumption for 8 weeks changes the proteome profile of kidney and heart tissue samples

297 of experimental rat models with metabolic syndrome, which has beneficial effects on the cardiovascular system, free radical scavenging and lipid and glucose metabolism.

In Project 1, we summarize by a systematic review the available scientific evidence of the effects of hesperidin consumption on cardiovascular risk biomarkers. Animal studies and human RCT published between January 2003 and January 2017 were included in the systematic review. Mice and rats and human subjects had at least one CVDRF, they participated in sustained and single dose studies and consumed a hesperidin dose through OJ or capsule; and orally, by gavage and intravenous administration in the case of animal models.

In animal studies, hesperidin consumption improves blood glucose, TC, LDL-c, and TG levels. No significant changes were observed in anthropometric parameters, BP, inflammation, or oxidative biomarkers. However, the sample size in some of the studies may be insufficient to obtain conclusive results. On the other hand, a conclusive inference cannot be drawn from the included RCT: most of the studies did not have the most appropriate population to evaluate the effects of different CVDRFs, and only 3 of the 11 included articles evaluated the diet of the subjects through a validated method and no one used consumption biomarkers. There may be factors that influence hesperidin bioavailability or other phenolic compounds of the diet that can influence the observed effects. Therefore, further RCT with higher quality are needed to clarify the effects of hesperidin on CVD biomarkers.

298

To the best of our knowledge, the present study is the first to systematically review the effects of hesperidin on CVD risk biomarkers in animal studies and human RCT. In this sense, no systematic review or meta-analysis about the effects of hesperidin on CVDs was published after our work. However, in 2020, a review has been published reporting the scientific evidence of hesperidin consumption on CVDRFs and its role in intestinal microbiota.(112) In this review the mechanism of action by which hesperidin can exert beneficial effects was discussed together with the results of in vivo, in vitro and human clinical trial studies. In this work, although they demonstrate the influence of the intestinal microbiota on the absorption and subsequent effects of hesperidin consumption, more animal and human studies are needed to clarify this relationship. In this sense, their conclusion about the actual evidence of the hesperidin effects is the same as what we reported in our systematic review.

As there is a need to realize more randomized and controlled clinical trials to confirm the effects of hesperidin on CVD risk biomarkers, the CITRUS study was carried out.

In Project 2, we assessed the sustained and single dose effects of hesperidin consumption in OJ and EOJ on BP levels and PP in pre- and stage 1 hypertensive subjects. After 12 weeks of daily hesperidin consumption in 500 mL of OJ (392 mg of hesperidin) and EOJ (670 mg of hesperidin), decreased SBP and PP levels were observed in a dose- dependent manner. Therefore, regular consumption of OJ, especially EOJ, could be a co-adjuvant tool for BP and PP management.

299

Similarly, an open-labelled randomized controlled trial that evaluated the clinical effects of hesperidin in metabolic syndrome subjects was recently published.(190) In this study, after 12 weeks of hesperidin powder consumption (1 g/day), significant reductions in SBP of -5.68 mm Hg and serum TG of 50.06 mg/dL were observed. Moreover, a randomized, double-blinded, placebo-controlled clinical trial also performed in patients with metabolic syndrome showed that the consumption of 500 mg/day hesperidin by 2 capsules for 12 weeks significantly decreased the levels of blood glucose, TG, SBP and TNFα.(191) Therefore, hesperidin can improve metabolic abnormalities and inflammatory status in subjects with metabolic syndrome. Moreover, hesperidin can exerts anti-hypertensive effects increasing NO production,(192) improving endothelium-dependent vasodilatation and improving potassium channel activity.(193)

Furthermore, in the CITRUS study the genes that were significantly and differently decreased after 12 weeks of hesperidin consumption in EOJ related to BP levels were PTX3 and NAMPT. PTX3 is a marker of inflammation activation and is increased in hypertensive subjects.(194) Additionally, NAMPT is implicated in inflammation since it is a stimulator of pro-inflammatory cytokines, and it is increased in subjects with prehypertension.(195) Therefore, NAMPT could be a marker of risk and damage in prehypertensive subjects. Accordingly, the decreased expression of both genes could be partially responsible for the decreased BP and PP levels after hesperidin consumption because their expression was correlated with BP and PP.

300

Therefore, the CITRUS study provides further evidence of the effects of hesperidin consumption on CVDRFs such as BP. However, more in- depth analyses are necessary to better understand how hesperidin can exert its beneficial effects and understand its mechanisms of action. In this sense, an omics science approach through transcriptomics and metabolomics analysis was performed in subjects with pre- and stage 1 hypertension from the CITRUS study.

In Project 3, additionally, we determined whether the sustained and single dose consumption of hesperidin in OJ and EOJ can change the transcriptome profile of PBMCs of pre- and stage 1 hypertensive subjects. According to our results, after 12 weeks of OJ and EOJ consumption, pro-inflammatory genes were downregulated. However, after EOJ consumption more pro-inflammatory genes were differentially and borderline downregulated than after the consumption of OJ (such as CXCL2, CCL3, TNF, IL1B and PTGS2). These differentially expressed genes after hesperidin consumption influence the inflammatory response and the communication between immune cells preventing atherosclerotic plaque formation, inflammation in obesity, short-term mortality and hypertension, thus decreasing the risk of developing CVDs. Additionally, the differential expression of a gene related to insulin resistance was only observed after the single dose consumption of hesperidin in OJ and EOJ. The OJ upregulated the DHRS9 gene compared to CD, increasing insulin resistance, but EOJ downregulated it compared to OJ. The different behaviour between the two hesperidin interventions could be due to the presence of fructose: fructose can increase the expression of DHRS9, increasing fructose metabolism,

301 plasma TG levels and insulin resistance. In contrast, after EOJ consumption, the gene expression of DHRS9 decreased, which could be because hesperidin can counteract with the fructose content since the evidence suggests that higher doses of hesperidin in rats improve the insulin resistance and decrease blood glucose levels.(143,144)

On the other hand, and contrary to expectations, hesperidin in OJ and EOJ also showed the capacity to downregulate anti-inflammatory related genes such as NFKB1A and TNFAIP3. The downregulated effect observed could also be explained by the fructose content of the fruit juices, although the product of the present RCT was natural juice or enriched juice. Some evidence suggests that a high consumption of fructose increases CVD risk because of its influence on inflammatory response.(196) However, other evidence is controversial because its pro-inflammatory effects are only observed after artificially sweetened beverage consumption but not after fruit juice consumption.(156)

The present work is the first study to realize a transcriptomic approach of PBMCs in subjects with pre- and stage 1 hypertension after sustained and single dose consumption of hesperidin in OJ and EOJ. Transcriptomic analysis offers a unique opportunity to determine the effect of bioactive compounds such as hesperidin on metabolic and biological pathways. Additionally, transcriptomic analysis provides a way of knowing how the regulation of some pathways impacts the progression of chronic diseases.(197) Moreover, the transcriptomic approach allows for the identification of genes as biomarkers.

Thus, the new information provided through our transcriptomic approach would allow us to better understand how hesperidin

302 consumption can influence cardiovascular health, providing an explanation of the observed health effects. However, although our results are promising, more studies are needed to corroborate this hypothesis to better understand the mechanism of action of hesperidin and its action on gene expression.

In Project 4, we evaluated the effects of a single dose and sustained consumption of hesperidin for 12 weeks in OJ and EOJ in subjects with pre- or stage 1 hypertension on plasma, serum and urinary metabolomic profiles through targeted and nontargeted approach. After 12 weeks of hesperidin consumption in OJ and EOJ, compared to CD, the targeted metabolomics approach showed that the 6 compliance biomarkers were significantly increased in urine and plasma, but hesperetin 7-O-B-D- glucuronide was the only metabolite that showed differences between OJ and EOJ, indicating that hesperetin 7-O-B-D-glucuronide could be a candidate marker to distinguish between different hesperidin doses consumed in the long-term. Additionally, we observed a negative correlation between the urine levels of hesperetin 7-O-B-D-glucuronide and SBP levels, suggesting that this metabolite can have hypotensive actions. To the best of our knowledge, this is the first study to show a negative correlation between the urine metabolite hesperetin 7-O-B-D- glucuronide and the levels of SBP.

On the other hand, nontargeted metabolomics approach showed that after hesperidin consumption in OJ and EOJ for 12 weeks and after 6 hours of a single dose study, serum and urine metabolites that are inversely associated with BP levels,(173) such as proline betaine, were significantly increased. Moreover, decreased serum and urine levels of

303 the metabolites valine, leucine and isoleucine were reported after OJ and EOJ consumption, which were related to lower BP levels and improvement in insulin resistance.(125) Additionally, metabolites such as NAG and pseudouridine, related to inflammatory and oxidative stress status,(176) were decreased in serum and urine, after 12 weeks of OJ and EOJ consumption. Also, hesperidin consumption in OJ and EOJ, changes the choline metabolism changing related metabolites such as trimethylamine and dimethylglycine, which are increased, suggesting a lower microbial choline metabolism and thereby alterations in one- carbon metabolism that could improve the SBP levels in these patients. Furthermore, uraemic toxins such as dimethylamine, 4-cresyl sulfate and creatinine were decreased in urine after OJ and EOJ consumption, which is beneficial for the cardiovascular system because their accumulation produces vascular inflammation and endothelial dysfunction.(183)

On the other hand, and as reported in the literature, we observed a large interindividual variability in flavanone absorption and for that reason we divided the subjects into two groups in the single dose study: high and low total flavanone absorbers regardless of their consumption of OJ and EOJ. In this sense, we observed that the subjects who absorbed more flavanones showed higher plasma levels of ketone bodies such as acetoacetate and 3-hydroxybutyrate 6 h hours postprandially. It is known that increased ketone bodies produced by a ketogenic diet produce vasodilation and lower BP levels.(187) Therefore, this could be another mechanism through which hesperidin can improve BP levels in mildly hypertensive subjects.

304

In Project 5, we evaluated the effects of hesperidin supplementation at 100 mg/kg body weight/day for 8 weeks on the proteomic profiles of heart and kidney tissue samples from healthy and metabolic syndrome rats. The human equivalent dose of 100 mg/body weigh/day hesperidin was 1350 mg per day for a 60 kg human,(198) a dose achievable with an orange juice enriched with hesperidin.

After 8 weeks of hesperidin supplementation by diet, the metabolic syndrome rats had changes in their proteome in both heart and kidney tissues. In the heart and kidney tissue samples, 35 proteins and 53 proteins were differentially expressed, respectively. To the best of our knowledge, there are no data about the capacity of hesperidin to change the proteome profile of tissue samples of metabolic syndrome rats. Therefore, this is the first study to observe it. According to our results, hesperidin supplementation showed positive effects on different parameters in metabolic syndrome rats, upregulating the expression of proteins related to the cardiovascular system (ATP1A1), BP and endothelial function (ATP5PF and IGFBP7, respectively), and atherosclerosis pathogenesis (TPT). Additionally, hesperidin supplementation exerts positive effects by downregulating the expression of proteins related to free radical scavenging (ALDH7A1), heart development (FLNA), glucose metabolism (CD2AP) and insulin sensitivity (EPHX2). However, hesperidin can also change the expression of some proteins with negative or controversial effects on cardiovascular health: hesperidin downregulated the expression of antioxidant proteins (SOD1 and TXN2), proteins related negatively to BP (SLC9A3R1) and proteins related negatively to lipid metabolism (APOA4).

305

In this sense, and contrary to expectations, we observed controversial results after the proteomics approach in rats as in the transcriptomics approach in humans. Therefore, more studies are needed in this area to explain and clarify the effects of hesperidin on these differentially expressed proteins. On the other hand, some studies have performed proteomics analysis in cells and rats treated with flavonoids such as quercetin and observed interesting and positive results. After quercetin treatment in K562 cells (a cellular model of human chronic myeloid leukaemia), several proteins related to RNA metabolism, the antioxidant defense system and lipid metabolism changed, influencing the early stages of the apoptosis response.(199) On the other hand, in rats with cerebral ischaemia, quercetin treatment can change the expression of proteins related to cellular differentiation, metabolism and oxidative stress, reducing ischaemic injury.(200)

Thus, promising effects of hesperidin capacity to influence the proteome profile to exert beneficial effects on interestingly expressed proteins to promote cardiovascular health were reported and need to be verify in humans.

OVERALL DISCUSSION

We can observe that higher doses of hesperidin (670 mg/day in 500 mL/day in EOJ) have more beneficial effects than usual doses (392 mg/day in 500 mL/day in OJ). This is suggested because of the results from the CITRUS study regarding BP and PP, the transcriptomic approach performed in PBMCs which observed interestingly actions

306 decreasing proinflammatory genes, and the metabolomic approach realised in urine and plasma/serum samples which suggests changes on several metabolites related with lower levels of BP and beneficial effects on cardiovascular system. Therefore, hesperidin consumption in OJ and EOJ can decrease BP levels and PP in a dose-dependent manner in subjects with pre- and stage 1 hypertension, and the mechanism of action of the flavonoid hesperidin, which exerts its beneficial effects, can be explained through transcriptomics and metabolomics analyses. Hesperidin showed the ability to decrease the expression of pro- inflammatory genes and higher doses can improve glucose metabolism by interfering with the expression of genes related to insulin resistance. Furthermore, hesperidin in OJ and EOJ has an impact on the serum metabolomic profile, decreasing endogenous metabolite levels related to BP and inflammation, and decreasing urinary excretion of uraemic toxins and metabolites related to oxidative stress. Moreover, an experimental study in rats with metabolic syndrome also showed the capacity of hesperidin to influence kidney and heart proteome profiles by proteomics analysis. In this sense, hesperidin can change the expression of proteins related to cardiovascular health, improving endothelial function and BP homeostasis. Additionally, hesperidin showed the capacity to change protein expression of heart and kidney tissues to exert antioxidant effects and improve of glucose metabolism in metabolic syndrome rats.

Thus, in the present work we reported promising effects of hesperidin consumption to promote cardiovascular health (Figure 14).

307

308

7. PERSPECTIVES

309

310

First, although our results are promising and indicate beneficial effects of hesperidin consumption reducing SBP linked to changes in gene expression, metabolites in humans and proteins in syndrome metabolic rats, more studies are needed to confirm our results.

Moreover, to assess the effects of hesperidin on other CVDRFs biomarkers such as lipid metabolism, endothelial function, glucose metabolism and anthropometric parameters, further research is needed.

Additionally, reviewing the types of studies performed with hesperidin consumption, the participants included in the RCT must be the proper population to ensure the observed positive effects on specific cardiovascular biomarkers. For example, if the hesperidin effects on TC plasma levels are to be evaluated, adult subjects with hypercholesterolaemia must be selected; or if the hesperidin effects on body weight want are to be evaluated, overweight or obese subjects must be selected.

On the other hand, there is a bidirectional relationship between phenolic compounds and the microbiota of the human gut since phenolic compounds can change the microbiota population (Figure 15).(56)

311

Figure 15. Flavonoids metabolism in the colon and metabolite absorption. 1. Dietary flavonoids can exert local effects in gastrointestinal tract, 2. Can interact with microbiota changing its profile and produce mainly aromatic and phenolic acids that will be absorbed producing systemic effects or exerting local effects; 3. and be metabolized by intestinal epithelial cells. Later metabolites will be absorbed or excreted. Abbreviations: GI, gastrointestinal. Source: (Fraga CG, et al. 2019).(56)

The microbiota is a complex ecosystem that depends on individual characteristics and their environmental conditions, and plays a very important role in health status by modulating the immune system and protecting against pathogenic microorganisms.(201) Therefore, the microbiota influences on the development of several diseases.

312

In this sense, in other human studies OJ consumption showed a prebiotic effect in the intestinal mucosa by decreasing pathogenic microorganisms and increasing positive effects on the intestine.(202) Moreover, hesperidin has demonstrated in vivo and in vitro the capacity to inhibit the growth of gram-positive bacteria, thereby inhibiting the progression of pathogenic bacteria, and the ability to promote the growth of beneficial bacteria, thus demonstrating immunomodulatory action on the gut.(203) Furthermore, regarding BP, a recent study that was carried using the same rats as in Project 5 of the present thesis evaluated the effects of hesperidin consumption on the microbiota in metabolic syndrome rats, showing that hesperidin supplementation alters microbiota by acting as a hypertension modulator and modifying protein expression related to pathways metabolism such as amino acids and energy metabolism.(204)

For that reason, since gut microbiota have a crucial role in disease development, such as CVDs, and also phenolic compounds is metabolized by gut microbiota, stool samples should also be collected in future RCT to evaluate the hesperidin effects on microbiota and their implication for CVDs. Furthermore, a metagenomic approach could be useful to understand the complex relationship between CVDs, the gut microbiome and hesperidin intake since metagenomics can analyse the genomic content of all microorganisms present in an ecosystem to define their biodiversity in each experimental condition.

On the other hand, to the best of our knowledge, no proteomics approach has been realized in humans to evaluate the effects of

313 hesperidin consumption on proteomic profiles. Therefore, blood samples can also be collected in RCT to determine changes in serum and plasma proteins and to relate them to cardiovascular health.

Finally, further analysis from the CITRUS study results will be realized to integrate all the information generated and to better relate the changes at the molecular level to the clinical changes observed. In this sense, more correlations between clinical parameters such as BP, PP, or endothelial function with inflammation biomarkers, differentially expressed genes and endogenous metabolites can be realized.

Second, a multi-omics approach will be proposed thanks to the omics approach undertaken in the present work through transcriptomics, metabolomics and proteomics analysis, it has been possible to know the impact of the flavanone hesperidin on the transcriptome, metabolome and proteome to understand how hesperidin exerts beneficial effects on the cardiovascular system and CVD biomarkers. As results, each omics science provides specific insight into one study factor; however, an integrative new analysis proposal is currently emerging in the world of science and bioinformatics, and it is called the “multi-omics” approach.

Multi-omics is an integrative analysis of omics data from different omics levels with the objective of better understanding their interrelation and combined influence on molecular function, disease aetiology and disease development.(129) However, the integrative analysis of different omics data is not straightforward and has several logistic and computational changes.(205) Nevertheless, realizing a multi-omics approach is one more step that future studies can perform

314 to integrate multi-omics profiles into the investigation of the mechanisms associated with CVDs.

Third, considering that hesperidin has the ability to act on multiple factors, such as BP, lipid and glucose metabolism, as well as on inflammation and oxidative stress, studies could be carried out to determine the effects of hesperidin consumption in other chronic diseases such as cancer. In this sense, there is literature that reported that hesperidin consumption has beneficial effects on oxidation and other biological pathways related to metastasis and better prognosis described in tumor cells.(206,207)

Figure 16 shows the role of hesperidin in apoptosis and the cell cycle: hesperidin generates reactive oxygen species in cancer cells, activates mitochondrial pathways inducing apoptosis, and arrests the cancer cell cycle.

Figure 16. Role of hesperidin on apoptosis and cell cycle. Source: (Aggarwal V et al. 2020).(207)

315

Finally, with the currently active worldwide Coronavirus Disease 2019 (COVID-19) pandemic, studies have been carried out linking the consumption of hesperidin with positive effects on COVID-19. The ability of hesperidin to inhibit angiotensin-converting enzyme 2, transmembrane protease serine 2 and binding immunoglobulin protein receptors is one of the reasons for its possible beneficial effects since these receptors are the most noticeable receptors causing COVID- 19.(208) Furthermore, flavonoids can have antiviral action because of their modulation of the immune system. Hesperidin showed the ability to bind to key proteins of severe acute respiratory syndrome coronavirus (SARS-CoV-2) with effective antiviral action that inhibited virus-induced cellular and systemic pathology.(209)

Therefore, there is evidence that supports the promising use of hesperidin in the prophylaxis and treatment of COVID-19: hesperidin can improve host cellular immunity against infection and anti- inflammatory action helping to control cytokine storms, and hesperidin mixed with specific drugs (diosmin co-administered with heparin) can protect against venous thromboembolism, which prevents the progression of the disease.

Figure 17 shows the possible effect of hesperidin on the prophylaxis and treatment of COVID-19.(210)

316

Figure 17. Possible effect of hesperidin on the prophylaxis and treatment of COVID-19. Abbreviations: ACE-2, angiotensin- converting enzyme 2. Source: (Haggag YA et al. 2020).(210)

In summary, it can be observed that there are future study routes in relation to hesperidin and its capacity to protect against chronic diseases that can be followed and continued. Further RCT with higher quality with the collection of stool samples for the metagenomics approach, the collection of plasma samples for the proteomics approach in humans, realizing a multi-omics analysis to integrate all the biological information and evaluate the effects of hesperidin consumption on other chronic diseases such as cancer, can be future goals in the world of nutrition and health for the prevention and treatment of diseases. Moreover, studies that relate hesperidin consumption and COVID-19 pathogenesis will be interesting given the possible applications it may have.

317

318

8. CONCLUSIONS

319

320

According to the results obtained, the conclusions about the effects of hesperidin consumption on CVD risk biomarkers are as follows:

-After the systematic review, hesperidin consumption was found to improve glucose levels and lipid profiles in animal models, but no definitive conclusion regarding the effects of hesperidin in humans can currently be drawn (Objective 1 and Project 1).

-After the CITRUS study, the intake of hesperidin in OJ decreased SBP and PP after sustained consumption in a dose-dependent manner with the hesperidin content of the beverage administered. Chronic consumption of EOJ enhances the postprandial response of decreasing SBP and PP. Decreases in homocysteine, uric acid and inflammatory markers at the systemic level and in PTX3 and NAMPT at the transcriptomic level could account for the observed changes in BP and PP (Objective 2 and Project 2).

-After the transcriptomic analysis from CITRUS study, the single dose consumption of higher doses of hesperidin could induce a better response than the consumption of the naturally occurring doses of hesperidin in OJ because of their improvement of insulin resistance. Moreover, the sustained consumption of hesperidin in EOJ decrease the expression of proinflammatory genes providing a possible mechanism of action on inflammation pathway and thereby could induces beneficial effects on the cardiovascular system. (Objective 3 and Project 3).

-The plasma and urine metabolite hesperetin 7-O-B-D-glucuronide is the only metabolite that reported hesperidin dose response differences,

321 and in urine it is inversely correlated with SBP levels. In addition, hesperidin consumption in OJ and EOJ for 12 weeks changed the serum metabolomic profile, decreasing the levels of endogenous metabolites related to BP, inflammation, and oxidative stress, and decreasing the urinary excretion of uraemic toxins. Additionally, after a single dose of hesperidin, changes in the serum levels of metabolites related to reduced BP levels and anti-inflammatory effects were observed (Objective 4 and Project 4).

-After proteomic analysis in metabolic syndrome rats, hesperidin supplementation changed the proteomic profiles of the heart and kidney tissues and had a beneficial impact on the cardiovascular system, free radical scavenging, and lipid and glucose metabolism (Objective 5 and Project 5).

OVERALL CONCLUSION

Finally, the overall conclusion is that hesperidin reduces human BP and PP in a dose-dependent manner. Thus, the hesperidin enrichment achieved with EOJ can be a useful co-adjuvant tool for BP and PP management in pre- and stage 1 hypertensive subjects. The mechanisms of action by which hesperidin exerts its beneficial effects can be explained through transcriptomics and metabolomics approaches in humans which demonstrated cardioprotective actions through decreases in pro-inflammatory genes (transcriptomic approach), decreases in serum endogenous metabolites related to BP and oxidative stress and decreases in the urinary excretion of uraemic toxins (metabolomic approach). Moreover, the proteomic approach realized in kidney and

322 heart tissues of metabolic syndrome rats, showed that hesperidin changes proteomic profiles exerting positive effects on two main organs involved on BP regulation and cardiovascular system.

323

324

9. REFERENCES

325

326

1. World Health Organization. WHO | Top 10 causes of death [Internet]. 2018 [cited 2020 Sep 24]. p. 1. Available from: https://www.who.int/gho/mortality_burden_disease/causes_dea th/top_10/en/ 2. Federation WH. Risk factors - World Heart Federation [Internet]. 2020 [cited 2020 Sep 24]. p. 1. Available from: https://www.world-heart-federation.org/resources/risk-factors/ 3. World Health Organization. WHO | Top 10 causes of death. WHO [Internet]. 2018 [cited 2018 Jun 5]; Available from: http://www.who.int/gho/mortality_burden_disease/causes_deat h/top_10/en/ 4. Toma L, Sanda GM, Niculescu LS, Deleanu M, Sima AV, Stancu CS. Phenolic compounds exerting lipid-regulatory, anti- inflammatory and epigenetic effects as complementary treatments in cardiovascular diseases. Biomolecules. 2020;10(4). 5. Vailati-Riboni, M; Palombo, V; Loor J. What are omics sciences? Springer Int Publ Switz. 2017; 6. Chareonrungrueangchai K, Wongkawinwoot K, Anothaisintawee T, Reutrakul S. Dietary factors and risks of cardiovascular diseases: An umbrella review. Nutrients. 2020;12(4). 7. Ditano-Vázquez P, Torres-Peña JD, Galeano-Valle F, Pérez- Caballero AI, Demelo-Rodríguez P, Lopez-Miranda J, et al. The fluid aspect of the mediterranean diet in the prevention and management of cardiovascular disease and diabetes: The role of polyphenol content in moderate consumption of wine and olive oil. Nutrients. 2019;11(11).

327

8. Filippou CD, Tsioufis CP, Thomopoulos CG, Mihas CC, Dimitriadis KS, Sotiropoulou LI, et al. Dietary Approaches to Stop Hypertension (DASH) Diet and Blood Pressure Reduction in Adults with and without Hypertension: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Adv Nutr. 2020;1–11. 9. Del Rio D, Rodriguez-Mateos A, Spencer JPE, Tognolini M, Borges G, Crozier A. Dietary (Poly)phenolics in Human Health: Structures, Bioavailability, and Evidence of Protective Effects Against Chronic Diseases. Antioxid Redox Signal. 2013; 10. Giglio RV, Patti AM, Cicero AFG, Lippi G, Rizzo M, Toth PP, et al. Polyphenols: Potential Use in the Prevention and Treatment of Cardiovascular Diseases. Curr Pharm Des. 2018;24(2):239–58. 11. Spieth PM, Kubasch AS, Isabel Penzlin A, Min-Woo Illigens B, Barlinn K, Siepmann T, et al. NDT-101938-randomized- clinical-trials--a-matter-of-design. 2016;12:1341–9. Available from: http://dx.doi.org/10.2147/NDT.S101938 12. Farràs M, Arranz S, Carrión S, Subirana I, Muñoz-Aguayo D, Blanchart G, et al. A functional virgin olive oil enriched with olive oil and thyme phenolic compounds improves the expression of cholesterol efflux-related genes: A randomized, crossover, controlled trial. Nutrients. 2019;11(8):1–10. 13. Pedret A, Catalán Ú, Fernández-Castillejo S, Farràs M, Valls RM, Rubió L, et al. Impact of virgin olive oil and phenol- enriched virgin olive oils on the HDL proteome in hypercholesterolemic subjects: A double blind, randomized,

328

controlled, cross-over clinical trial (VOHF study). PLoS One. 2015;10(6):1–19. 14. Li C, Schluesener H. Health-promoting effects of the citrus flavanone hesperidin. Crit Rev Food Sci Nutr. 2017;57(3):613– 31. 15. Yamamoto M, Jokura H, Hashizume K, Ominami H, Shibuya Y, Suzuki A, et al. Hesperidin metabolite hesperetin-7-O- glucuronide, but not hesperetin-3′-O-glucuronide, exerts hypotensive, vasodilatory, and anti-inflammatory activities. Food Funct [Internet]. 2013;4(9):1346. Available from: http://xlink.rsc.org/?DOI=c3fo60030k 16. Yamamoto M, Suzuki a, Hase T. Short-term effects of glucosyl hesperidin and hesperetin on blood pressure and vascular endothelial function in spontaneously hypertensive rats. J Nutr Sci Vitaminol. 2008; 17. Zamora-Ros R, Andres-Lacueva C, Lamuela-Raventós RM, Berenguer T, Jakszyn P, Barricarte A, et al. Estimation of Dietary Sources and Flavonoid Intake in a Spanish Adult Population (EPIC-Spain). J Am Diet Assoc [Internet]. 2010;110(3):390–8. Available from: http://dx.doi.org/10.1016/j.jada.2009.11.024 18. Zheng J, Zhou Y, Li S, Zhang P, Zhou T, Xu DP, et al. Effects and mechanisms of fruit and vegetable juices on cardiovascular diseases. Int J Mol Sci. 2017;18(3). 19. Estadística IN de. Evolución de la esperanza de vida al nacimiento por periodo y sexo. Brecha de género. España. [Internet]. 2019 [cited 2020 Sep 24]. p. 1. Available from: https://www.ine.es/jaxi/Datos.htm?path=/t00/mujeres_hombres

329

/tablas_1/l0/&file=d01001.px#!tabs-tabla 20. World Health Organization. WHO | Life expectancy [Internet]. 2020 [cited 2020 Sep 24]. p. 1. Available from: https://www.who.int/gho/mortality_burden_disease/life_tables/ situation_trends_text/en/ 21. North, Brian J; Sinclair DA. The intersection between aging and cardiovascular disease. Circ Res. 2012;8:1097–108. 22. Estadística IN de. Causas de muerte en España. 2018. 2019; 23. World Health Organization. WHO | About cardiovascular diseases [Internet]. 2020 [cited 2020 Sep 24]. p. 1. Available from: https://www.who.int/cardiovascular_diseases/about_cvd/en/ 24. Casas R, Castro-Barquero S, Estruch R, Sacanella E. Nutrition and cardiovascular health. Vol. 19, International Journal of Molecular Sciences. 2018. 1–31 p. 25. Katakami N. Mechanism of development of atherosclerosis and cardiovascular disease in diabetes mellitus. J Atheroscler Thromb. 2018;25(1):27–39. 26. Aboonabi A, Meyer RR, Singh I. The association between metabolic syndrome components and the development of atherosclerosis. J Hum Hypertens [Internet]. 2019;33(12):844– 55. Available from: http://dx.doi.org/10.1038/s41371-019- 0273-0 27. Herrero-Fernandez B, Gomez-Bris R, Somovilla-Crespo B, Gonzalez-Granado JM. Immunobiology of atherosclerosis: A complex net of interactions. Int J Mol Sci. 2019;20(21). 28. O’Donnell CJ, Elosua R. Cardiovascular risk factors. Insights from framingham heart study. Rev Esp Cardiol.

330

2008;61(3):299–310. 29. Kondo T, Nakano Y, Adachi S, Murohara T. Effects of Tobacco Smoking on Cardiovascular Disease. 2019;83(October):1980–5. 30. Francula-Zaninovic S, Nola IA. Management of Measurable Variable Cardiovascular Disease’ Risk Factors. Curr Cardiol Rev. 2018;14(3):153–63. 31. Mozaffarian D. Dietary and Policy Prioritites for CVD, diabetes and obesity - a comprehensive RV. Circulation. 2016;133(2):187–225. 32. Mach F, Baigent C, Catapano AL, Koskinas KC, Casula M, Badimon L, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: Lipid modification to reduce cardiovascular risk. Eur Heart J. 2020;41(1):111–88. 33. Perk J, De Backer G, Gohlke H, Graham I, Reiner Ž, Verschuren WMM, et al. European Guidelines on cardiovascular disease prevention in clinical practice (version 2012). The Fifth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by re. Atherosclerosis. 2012;223(1):1–68. 34. Piepoli MF, Hoes AW, Agewall S, Albus C, Brotons C, Catapano AL, et al. 2016 European Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2016;37(29):2315–81. 35. Piepoli MF, Hoes AW, Agewall S, Albus C, Brotons C, Catapano AL, et al. 2016 European Guidelines on cardiovascular disease prevention in clinical practice. Eur

331

Heart J. 2016;37(29):2315–81. 36. Lacroix S, Cantin J, Nigam A. Contemporary issues regarding nutrition in cardiovascular rehabilitation. Ann Phys Rehabil Med. 2017;60(1):36–42. 37. Cordain L, Eaton SB, Sebastian A, Mann N, Lindeberg S, Watkins BA, et al. Origins and evolution of the Western diet : health implications for the. 2005; 38. Zampelas A, Magriplis E. Dietary patterns and risk of cardiovascular diseases: A review of the evidence. Proc Nutr Soc. 2020;79(1):68–75. 39. Estruch R, Ros E, Salas-Salvadó J, Covas M-I, Corella D, Arós F, et al. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med. 2013; 40. Ravera A, Carubelli V, Sciatti E, Bonadei I, Gorga E, Cani D, et al. Nutrition and cardiovascular disease: Finding the perfect recipe for cardiovascular health. Nutrients. 2016;8(6). 41. Korakas E, Dimitriadis G, Raptis A, Lambadiari V. Dietary composition and cardiovascular risk: A mediator or a Bystander? Nutrients. 2018;10(12). 42. Afshin A, Micha R, Khatibzadeh S, Mozaffarian D. Consumption of nuts and legumes and risk of incident ischemic heart disease, stroke, and diabetes: A systematic review and meta-analysis. Am J Clin Nutr. 2014;100(1):278– 88. 43. Widmer RJ, Flammer AJ, Lerman LO, Lerman A. The Mediterranean diet, its components, and cardiovascular disease. Am J Med. 2015;128(3):229–38. 44. Chiavaroli L, Viguiliouk E, Nishi SK, Mejia SB, Rahelić D,

332

Kahleová H, et al. DASH dietary pattern and cardiometabolic outcomes: An umbrella review of systematic reviews and meta-analyses. Nutrients. 2019;11(2). 45. Kahleova H, Levin S, Barnard ND. Vegetarian Dietary Patterns and Cardiovascular Disease. Prog Cardiovasc Dis [Internet]. 2018;61(1):54–61. Available from: https://doi.org/10.1016/j.pcad.2018.05.002 46. Esselstyn CB, Gendy G, Doyle J, Golubic M, Roizen MF. A way to Reduce CAD? J Fam Pract [Internet]. 2014;3(7):356– 64. Available from: https://mdedge-files-live.s3.us-east- 2.amazonaws.com/files/s3fs-public/Document/September- 2017/JFP_06307_Article1.pdf 47. D RCSWSKFCHeAWLJOsHFASBNGFSGTK-TKDMJoDE. Association of Dietary, Circulating, and Supplement Fatty Acids with Coronary Risk. Ann Intern Med. 2014;160:398– 406. 48. Nakao T, Morita H. Omega-3 polyunsaturated fatty acid and coronary atherosclerotic plaques. Int Heart J. 2017;58(6):837– 9. 49. Eckel RH, Jakicic JM, Ard JD, De Jesus JM, Houston Miller N, Hubbard VS, et al. 2013 AHA/ACC guideline on lifestyle management to reduce cardiovascular risk: A report of the American College of cardiology/American Heart Association task force on practice guidelines. Circulation. 2014;129(25 SUPPL. 1):76–99. 50. Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/

333

ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol [Internet]. 2019;73(24):3168–209. Available from: http://dx.doi.org/10.1016/j.jacc.2018.11.002 51. Mohammadifard N, Gotay C, Humphries KH, Ignaszewski A, Esmaillzadeh A, Sarrafzadegan N. Electrolyte minerals intake and cardiovascular health. Crit Rev Food Sci Nutr [Internet]. 2019;59(15):2375–85. Available from: https://doi.org/10.1080/10408398.2018.1453474 52. Asgary S, Rastqar A, Keshvari M. Functional Food and Cardiovascular Disease Prevention and Treatment: A Review. J Am Coll Nutr [Internet]. 2018;37(5):429–55. Available from: https://doi.org/10.1080/07315724.2017.1410867 53. Tufarelli V, Casalino E, D’Alessandro AG, Laudadio V. Dietary Phenolic Compounds: Biochemistry, Metabolism and Significance in Animal and Human Health. Curr Drug Metab. 2017; 54. Medina-Remón A, Zamora-Ros R, Rotchés-Ribalta M, Andres-Lacueva C, Martínez-González MA, Covas MI, et al. Total polyphenol excretion and blood pressure in subjects at high cardiovascular risk. Nutr Metab Cardiovasc Dis. 2011;21(5):323–31. 55. Pedret A, Valls RM, Fernández-Castillejo S, Catalán Ú, Romeu M, Giralt M, et al. Polyphenol-rich foods exhibit DNA antioxidative properties and protect the glutathione system in healthy subjects. Mol Nutr Food Res [Internet].

334

2012;56(7):1025–33. Available from: http://www.ncbi.nlm.nih.gov/pubmed/22760977 56. Fraga CG, Croft KD, Kennedy DO, Tomás-Barberán FA. The effects of polyphenols and other bioactives on human health. Food Funct. 2019;10(2):514–28. 57. Ivey KL, Jensen MK, Hodgson JM, Eliassen AH, Cassidy A, Rimm EB. Association of flavonoid-rich foods and flavonoids with risk of all-cause mortality. Br J Nutr. 2017;117(10):1470– 7. 58. Bondonno CP, Croft KD, Ward N, Considine MJ, Hodgson JM. Dietary flavonoids and nitrate: Effects on nitric oxide and vascular function. Nutr Rev. 2015;73(4):216–35. 59. Perez-vizcaino F, Fraga CG. Research trends in fl avonoids and health. Arch Biochem Biophys [Internet]. 2018;646(March):107–12. Available from: https://doi.org/10.1016/j.abb.2018.03.022 60. Rees A, Dodd GF, Spencer JPE. The effects of flavonoids on cardiovascular health: A review of human intervention trials and implications for cerebrovascular function. Nutrients. 2018;10(12). 61. Trial C, Fernández-castillejo S, Solà R. Phenol-enriched olive oils modify paraoxonase – related variables : a. :1–39. 62. Bondonno NP, Bondonno CP, Blekkenhorst LC, Considine MJ, Maghzal G, Stocker R, et al. Flavonoid-Rich Apple Improves Endothelial Function in Individuals at Risk for Cardiovascular Disease: A Randomized Controlled Clinical Trial. Mol Nutr Food Res. 2018;62(3):1–33. 63. Mulvihill EE, Burke AC, Huff MW. Citrus Flavonoids as

335

Regulators of Lipoprotein Metabolism and Atherosclerosis. Annu Rev Nutr. 2016;36:275–99. 64. Vazhappilly CG, Ansari SA, Al-Jaleeli R, Al-Azawi AM, Ramadan WS, Menon V, et al. Role of flavonoids in thrombotic, cardiovascular, and inflammatory diseases. Inflammopharmacology [Internet]. 2019;27(5):863–9. Available from: https://doi.org/10.1007/s10787-019-00612-6 65. Muller CJF, Malherbe CJ, Chellan N, Yagasaki K, Miura Y, Joubert E. Potential of rooibos, its major C-glucosyl flavonoids, and Z-2-(β-D-glucopyranosyloxy)-3- phenylpropenoic acid in prevention of metabolic syndrome. Crit Rev Food Sci Nutr. 2018;58(2):227–46. 66. Cassidy A, Bertoia M, Chiuve S, Flint A, Forman J, Rimm EB. Habitual intake of anthocyanins and flavanones and risk of cardiovascular disease in men. Am J Clin Nutr. 2016;104(3):587–94. 67. Jacques PF, Cassidy A, Rogers G, Peterson JJ, Meigs JB, Dwyer JT. Higher dietary flavonol intake is associated with lower incidence of type 2 diabetes. J Nutr. 2013;143(9):1474– 80. 68. Leyva-Soto A, Chavez-Santoscoy RA, Lara-Jacobo LR, Chavez-Santoscoy AV, Gonzalez-Cobian LN. Daily consumption of chocolate rich in flavonoids decreases cellular genotoxicity and improves biochemical parameters of lipid and glucose metabolism. Molecules. 2018;23(9):1–12. 69. Johnson CH, Ivanisevic J, Siuzdak G. Metabolomics: Beyond biomarkers and towards mechanisms. Nat Rev Mol Cell Biol. 2016;17(7):451–9.

336

70. Gutiérrez-Venegas G. Flavonoides en el tratamiento de la hipertensión en pacientes geriátricos. Rev Med Inst Mex Seguro Soc [Internet]. 2018;56(1):94–101. Available from: https://www.medigraphic.com/pdfs/imss/im-2018/ims181l.pdf 71. Morand C, Dubray C, Milenkovic D, Lioger D, Martin JF, Scalbert A, et al. Hesperidin contributes to the vascular protective effects of orange juice: A randomized crossover study in healthy volunteers. Am J Clin Nutr. 2011;93(1):73– 80. 72. Kim HS, Quon MJ, Kim J a. New insights into the mechanisms of polyphenols beyond antioxidant properties; lessons from the green tea polyphenol, epigallocatechin 3-gallate. Redox Biol [Internet]. 2014;2(1):187–95. Available from: http://dx.doi.org/10.1016/j.redox.2013.12.022 73. Yahfoufi N, Alsadi N, Jambi M, Matar C. The immunomodulatory and anti-inflammatory role of polyphenols. Nutrients. 2018;10(11):1–23. 74. Dariush Mozaffarian JHW. Flavonoids, Dairy Foods, and Cardiovascular and Metabolic Health: A Review of Emerging Biologic Pathways. Circ Res. 2018;122(2):369–84. 75. Rafiq S, Kaul R, Sofi SA, Bashir N, Nazir F, Ahmad Nayik G. Citrus peel as a source of functional ingredient: A review. J Saudi Soc Agric Sci. 2016; 76. Hajialyani M, Farzaei MH, Echeverría J, Nabavi SM, Uriarte E, Eduardo SS. Hesperidin as a neuroprotective agent: A review of animal and clinical evidence. Molecules. 2019;24(3). 77. Xiong H, Wang J, Ran Q, Lou G, Peng C, Gan Q, et al. Hesperidin: A therapeutic agent for obesity. Drug Des Devel

337

Ther. 2019;13:3855–66. 78. Dauchet L, Ferrières J, Arveiler D, Yarnell JW, Gey F, Ducimetière P, et al. Frequency of fruit and vegetable consumption and coronary heart disease in France and Northern Ireland: the PRIME study. Br J Nutr. 2004; 79. Li C, Schluesener H. Health-promoting effects of the citrus flavanone hesperidin. Crit Rev Food Sci Nutr. 2017;57(3):613– 31. 80. Li C, Schluesener H. Health-promoting effects of the citrus flavanone hesperidin. Crit Rev Food Sci Nutr. 2017; 81. Sun YZ, Chen JF, Shen LM, Zhou J, Wang CF. Anti- atherosclerotic effect of hesperidin in LDLr−/−mice and its possible mechanism. Eur J Pharmacol [Internet]. 2017;815(August):109–17. Available from: http://dx.doi.org/10.1016/j.ejphar.2017.09.010 82. Jung UJ, Lee M-K, Jeong K-S, Choi M-S. The hypoglycemic effects of hesperidin and naringin are partly mediated by hepatic glucose-regulating enzymes in C57BL/KsJ-db/db mice. J Nutr. 2004;134(10):2499–503. 83. Miwa Y, Mitsuzumi H, Sunayama T, Yamada M, Okada K, Kubota M, et al. Glucosyl hesperidin lowers serum triglyceride level in hypertriglyceridemic subjects through the improvement of very low-density lipoprotein metabolic abnormality. J Nutr Sci Vitaminol (Tokyo) [Internet]. 2005;51(6):460–70. Available from: http://www.ncbi.nlm.nih.gov/pubmed/16521708 84. Jung UJ, Lee MK, Park YB, Kang MA, Choi MS. Effect of citrus flavonoids on lipid metabolism and glucose-regulating

338

enzyme mRNA levels in type-2 diabetic mice. Int J Biochem Cell Biol. 2006;38(7):1134–45. 85. Kurowska EM, Spence JD, Jordan J, Wetmore S, Freeman DJ, Piche LA, et al. HDL-cholesterol-raising effect of orange juice in subjects with hypercholesterolemia. Am J Clin Nutr. 2000;72(5):1095–100. 86. Haidari F, Heybar H, Jalali MT, Ahmadi Engali K, Helli B, Shirbeigi E. Hesperidin Supplementation Modulates Inflammatory Responses Following Myocardial Infarction. J Am Coll Nutr. 2015;34(3):205–11. 87. Selvaraj P, Pugalendi KV. Hesperidin, a flavanone glycoside, on lipid peroxidation and antioxidant status in experimental myocardial ischemic rats. Redox Rep. 2010;15(5):217–23. 88. Mohammadi M, Ramezani-Jolfaie N, Lorzadeh E, Khoshbakht Y, Salehi-Abargouei A. Hesperidin, a major flavonoid in orange juice, might not affect lipid profile and blood pressure: A systematic review and meta-analysis of randomized controlled clinical trials. Phyther Res. 2019;33(3):534–45. 89. Demonty I, Lin Y, Zebregs YEMP, Vermeer MA, van der Knaap HCM, Jakel M, et al. The Citrus Flavonoids Hesperidin and Naringin Do Not Affect Serum Cholesterol in Moderately Hypercholesterolemic Men and Women. J Nutr [Internet]. 2010;140(9):1615–20. Available from: http://jn.nutrition.org/cgi/doi/10.3945/jn.110.124735 90. Mosqueda-Solís; A, Arrate Lasa; Gómez-Zorita S. Screening of potencial anti-adipogenic effects of phenoli compounds showing differical chemical structure in 3T3-L1 preadipocytes. Food Funct. 2017;8(10):3576–86.

339

91. Mosqueda-Solís A, Sánchez J, Portillo MP, Palou A, Picó C. Combination of capsaicin and hesperidin reduces the effectiveness of each compound to decrease the adipocyte size and to induce browning features in adipose tissue of western diet fed rats. J Agric Food Chem. 2018;66(37):9679–89. 92. Kim JK, Jeong HW, Kim AY, Hong YD, Lee JH, Choi JK, et al. Green satsuma mandarin orange (Citrus unshiu) extract reduces adiposity and induces uncoupling protein expression in skeletal muscle of obese mice. Food Sci Biotechnol [Internet]. 2019;28(3):873–9. Available from: https://doi.org/10.1007/s10068-018-0503-1 93. Aptekmann NP, Cesar TB. Orange juice improved lipid profile and blood lactate of overweight middle-aged women subjected to aerobic training. Maturitas. 2010; 94. Rangel-Huerta OD, Aguilera CM, Martin M V., Soto MJ, Rico MC, Vallejo F, et al. Normal or High Polyphenol Concentration in Orange Juice Affects Antioxidant Activity, Blood Pressure, and Body Weight in Obese or Overweight Adults. J Nutr [Internet]. 2015;145(8):1808–16. Available from: http://jn.nutrition.org/cgi/doi/10.3945/jn.115.213660 95. Ribeiro C, Dourado G, Cesar T. Orange juice allied to a reduced-calorie diet results in weight loss and ameliorates obesity-related biomarkers: A randomized controlled trial. Nutrition [Internet]. 2017;38:13–9. Available from: http://dx.doi.org/10.1016/j.nut.2016.12.020 96. Simpson EJ, Mendis B, Macdonald IA. Orange juice consumption and its effect on blood lipid profile and indices of the metabolic syndrome; A randomised, controlled trial in an

340

at-risk population. Food Funct. 2016;7(4):1884–91. 97. Michael A. GJ, Guillermo G-C. Endothelial cell dysfunction and the pathobiology of atherosclerosis. Circ Res. 2016;176(1):139–48. 98. Wunpathe C, Potue P, Maneesai P, Bunbupha S, Prachaney P, Kukongviriyapan U, et al. Hesperidin Suppresses Renin- Angiotensin System Mediated NOX2 Over-Expression and Sympathoexcitation in 2K-1C Hypertensive Rats. Am J Chin Med. 2018;46(4):1–17. 99. Maneesai P, Bunbupha S, Potue P, Berkban T, Kukongviriyapan U, Kukongviriyapan V, et al. Hesperidin prevents nitric oxide deficiency-induced cardiovascular remodeling in rats via suppressing TGF-β1 and MMPsprotein expression. Nutrients. 2018;10(10):1–15. 100. Rangel-Huerta OD, Aguilera CM, Martin M V., Soto MJ, Rico MC, Vallejo F, et al. Normal or high polyphenol concentration in orange juice affects antioxidant activity, blood pressure, and body weight in obese or overweight adults. J Nutr. 2015;145(8):1808–16. 101. Rizza S, Muniyappa R, Iantorno M, Kim JA, Chen H, Pullikotil P, et al. Citrus polyphenol hesperidin stimulates production of nitric oxide in endothelial cells while improving endothelial function and reducing inflammatory markers in patients with metabolic syndrome. J Clin Endocrinol Metab. 2011;96(5):782–92. 102. Buscemi S, Rosafio G, Arcoleo G, Mattina A, Canino B, Montana M, et al. Effects of red orange juice intake on endothelial function and inflammatory markers in adult

341

subjects with increased. Am J Clin Nutr. 2012;95:1089–95. 103. Salden BN, Troost FJ, de Groot E, Stevens YR, Garcés-Rimón M, Possemiers S, et al. Randomized clinical trial on the efficacy of hesperidin 2S on validated cardiovascular biomarkers in healthy overweight individuals. Am J Clin Nutr [Internet]. 2016;104(6):1523–33. Available from: https://academic.oup.com/ajcn/article/104/6/1523/4564674 104. Dhanya R, Jayamurthy P. In vitro evaluation of antidiabetic potential of hesperidin and its aglycone hesperetin under oxidative stress in skeletal muscle cell line. Cell Biochem Funct. 2020;38(4):419–27. 105. Jung UJ, Lee MK, Jeong KS, Choi MS. The hypoglycemic effects of hesperidin and naringin are partly mediated by hepatic glucose-regulating enzymes in C57BL/KsJ-db/db mice. J Nutr. 2004;134(10):2499–503. 106. Jung UJ, Lee MK, Park YB, Kang MA, Choi MS. Effect of citrus flavonoids on lipid metabolism and glucose-regulating enzyme mRNA levels in type-2 diabetic mice. Int J Biochem Cell Biol. 2006; 107. Morand C, Dubray C, Milenkovic D, Lioger D, Franc J, Scalbert A. Hesperidin contributes to the vascular protective effects of orange juice : a randomized crossover study in healthy volunteers 1 – 3. Am J Clin Nutr. 2011;93(7):73–80. 108. Del Rio D, Rodriguez-Mateos A, Spencer JPE, Tognolini M, Borges G, Crozier A. Dietary (Poly)phenolics in Human Health: Structures, Bioavailability, and Evidence of Protective Effects Against Chronic Diseases. Antioxid Redox Signal [Internet]. 2013;18(14):1818–92. Available from:

342

http://online.liebertpub.com/doi/abs/10.1089/ars.2012.4581 109. Bredsdorff L, Nielsen ILF, Rasmussen SE, Cornett C, Barron D, Bouisset F, et al. Absorption, conjugation and excretion of the flavanones, naringenin and hesperetin from-rhamnosidase- treated orange juice in human subjects. Br J Nutr. 2010;103(11):1602–9. 110. Pereira-Caro G, Polyviou T, Ludwig IA, Nastase AM, Moreno- Rojas JM, Garcia AL, et al. Bioavailability of orange juice (poly)phenols: The impact of short-term cessation of training by male endurance athletes. Am J Clin Nutr. 2017;106(3):791– 800. 111. Stevens Y, Van Rymenant E, Grootaert C, Van Camp J, Possemiers S, Masclee A, et al. The intestinal fate of citrus flavanones and their effects on gastrointestinal health. Nutrients. 2019;11(7). 112. Mas-Capdevila A, Teichenne J, Domenech-Coca C, Caimari A, Bas JMD, Escoté X, et al. Effect of hesperidin on cardiovascular disease risk factors: The role of intestinal microbiota on hesperidin bioavailability. Nutrients. 2020;12(5). 113. Julian K. Aschoff, Ken M. Riedl, Jessica L. Cooperstone, Josef Högel, Anja Bosy-Westphal, Steven J. Schwartz, Reinhold Carle RMS. Urinary excretion of Citrus flavanones and their major catabolites after consumption of fresh oranges and pasteurized orange juice: A randomized cross-over study. Mol Nutr Food Res. 2016;60(12):2602–10. 114. Jiang Z, Zhou X, Li R, Michal JJ, Zhang S, Dodson M V., et al. Whole transcriptome analysis with sequencing: Methods, challenges and potential solutions. Cellular and Molecular Life

343

Sciences. 2015. 115. Robert W. McGarrah, MD; Scott B Crown, PhD; Goufang Zhang, PhD; Scati H Shah, MD; Cristopher B Newgard P. Cardiovascular metabolomics. Circ Res. 2018;122(9):1238–58. 116. M B. Transcriptome analysis. Intech Open [Internet]. 2013;i(tourism):13. Available from: http://dx.doi.org/10.1039/C7RA00172J%0Ahttps://www.intec hopen.com/books/advanced-biometric-technologies/liveness- detection-in- biometrics%0Ahttp://dx.doi.org/10.1016/j.colsurfa.2011.12.01 4 117. Aslam B, Basit M, Nisar MA, Khurshid M, Rasool MH. Proteomics: Technologies and their applications. J Chromatogr Sci. 2017;55(2):182–96. 118. Zhang Z, Wu S, Stenoien DL, Paša-Tolić L. High-throughput proteomics. Annu Rev Anal Chem. 2014;7:427–54. 119. Yates JR. A century of mass spectrometry: From atoms to proteomes. Nat Methods [Internet]. 2011;8(8):633–7. Available from: http://dx.doi.org/10.1038/nmeth.1659 120. Egerstedt A, Berntsson J, Smith ML, Gidlöf O, Nilsson R, Benson M, et al. Profiling of the plasma proteome across different stages of human heart failure. Nat Commun [Internet]. 2019;10(1):1–13. Available from: http://dx.doi.org/10.1038/s41467-019-13306-y 121. Prasatthong P, Meephat S, Rattanakanokchai S, Bunbupha S, Prachaney P, Maneesai P, et al. Hesperidin ameliorates signs of the metabolic syndrome and cardiac dysfunction via IRS/Akt/GLUT4 signaling pathway in a rat model of diet-

344

induced metabolic syndrome. Eur J Nutr [Internet]. 2020;(0123456789). Available from: https://doi.org/10.1007/s00394-020-02291-4 122. Newgard CB. Metabolomics and Metabolic Diseases: Where Do We Stand? Cell Metab. 2017;25(1):43–56. 123. Ussher JR, Elmariah S, Gerszten RE, Dyck JRB. The Emerging Role of Metabolomics in the Diagnosis and Prognosis of Cardiovascular Disease. J Am Coll Cardiol. 2016;68(25):2850–70. 124. Deirdre K. Tobias, Patrick R. Lawler, Paulo H. Harada, Olga V. Demler, Paul M Ridker, JoAnn E. Manson SC, Mora and S. Cardiovascular Disease in a Prospective Cohort of US Women. Circ Genom Precis Med. 2019;11(4):1–17. 125. Yoon MS. The emerging role of branched-chain amino acids in insulin resistance and metabolism. Nutrients. 2016;8(7). 126. Susan Cheng, MD, MMSc, MPH [Chair], Svati H. Shah, MD, MHS, FAHA [Co-Chair], Elizabeth J. Corwin, PhD, Oliver Fiehn, MD, Robert L. Fitzgerald, PhD, Robert E. Gerszten, MD, FAHA, Thomas Illig, PhD, Eugene P. Rhee, MD, Pothur R. Srinivas, PhD, MPH, Thomas J P on behalf of the AHAC on FG and TBC on C and SNC on CC and SC. Potential Impact and Study Considerations of Metabolomics in Cardiovascular Health and Disease A Scientific Statement From the American Heart Association. Circ Cardiovasc Genet. 2017;10(2). 127. Snyder KJKMP. Integrative omcis for health and diease. Nat Rev Genet. 2018;19(5):299–310. 128. Cifuentes A. Food analysis and foodomics. J Chromatogr A.

345

2009;1216(43):7109. 129. Misra BB, , Carl Langefeld MO, Cox LA. Integrated Omics: Tools, Advances, and Future Approaches. J Mol Endocrinol. 2018;JME(18):18–0055. 130. Pla-Paga L, Companys J, Calderon-Perez L, Llaurado E, Sola R, Valls RM, et al. Effects of hesperidin consumption on cardiovascular risk biomarkers: A systematic review of animal studies and human randomized clinical trials. Nutr Rev. 2019;77(12):1–20. 131. Homayouni F, Haidari F, Hedayati M, Zakerkish M, Ahmadi K. Blood pressure lowering and anti-inflammatory effects of hesperidin in type 2 diabetes; a randomized double-blind controlled clinical trial. Phyther Res. 2018;32(6):1073–9. 132. Shen J, Wilmot KA, Ghasemzadeh N, Molloy DL, Burkman G, Mekonnen G, et al. Mediterranean Dietary Patterns and Cardiovascular Health. Annu Rev Nutr. 2015; 133. Valls RM, Pedret A, Calderón-Pérez L, Llauradó E, Pla-Pagà L, Companys J, et al. Effects of hesperidin in orange juice on blood and pulse pressures in mildly hypertensive individuals: a randomized controlled trial (Citrus study). Eur J Nutr [Internet]. 2020;(Ldl). Available from: https://doi.org/10.1007/s00394-020-02279-0 134. Ahmadi A, Shadboorestan A. Oxidative stress and cancer; the role of hesperidin, a citrus natural bioflavonoid, as a cancer chemoprotective agent. Nutr Cancer. 2016;68(1):29–39. 135. Bordoni L, Gabbianelli R. Primers on nutrigenetics and nutri(epi)genomics: Origins and development of precision nutrition. Biochimie [Internet]. 2019;160:156–71. Available

346

from: https://doi.org/10.1016/j.biochi.2019.03.006 136. D’Amore S, Vacca M, Cariello M, Graziano G, D’Orazio A, Salvia R, et al. Genes and miRNA expression signatures in peripheral blood mononuclear cells in healthy subjects and patients with metabolic syndrome after acute intake of extra virgin olive oil. Biochim Biophys Acta - Mol Cell Biol Lipids [Internet]. 2016;1861(11):1671–80. Available from: http://dx.doi.org/10.1016/j.bbalip.2016.07.003 137. Pollack RM, Barzilai N, Anghel V, Kulkarni AS, Golden A, O’Broin P, et al. Resveratrol Improves Vascular Function and Mitochondrial Number but Not Glucose Metabolism in Older Adults. Journals Gerontol - Ser A Biol Sci Med Sci. 2017;72(12):1703–9. 138. Milenkovic D, Deval C, Dubray C, Mazur A, Morand C. Hesperidin displays relevant role in the nutrigenomic effect of orange juice on blood leukocytes in human volunteers: A randomized controlled Cross-Over study. PLoS One. 2011;6(11):1–9. 139. Cappuzzello C, Napolitano M, Arcelli D, Melillo G, Melchionna R, Di Vito L, et al. Gene expression profiles in peripheral blood mononuclear cells of chronic heart failure patients. Physiol Genomics. 2009;38(3):233–40. 140. D’Amore S, Vacca M, Graziano G, D’Orazio A, Cariello M, Martelli N, et al. Nuclear receptors expression chart in peripheral blood mononuclear cells identifies patients with Metabolic Syndrome. Biochim Biophys Acta - Mol Basis Dis. 2013;1832(12):2289–301. 141. Livak KJ, Schmittgen TD. Analysis of relative gene expression

347

data using real-time quantitative PCR and the 2-ΔΔCT method. Methods. 2001;25(4):402–8. 142. Vasanti S Malik, Hu FB. Fructose and Cardiometabolic Health: What the Evidence from Sugar-Sweetened Beverages Tells Us. J Am Coll Cardiol. 2015;66(14):1615–24. 143. Sundaram R, Nandhakumar E, Haseena Banu H. Hesperidin, a citrus flavonoid ameliorates hyperglycemia by regulating key enzymes of carbohydrate metabolism in streptozotocin-induced diabetic rats. Toxicol Mech Methods [Internet]. 2019;29(9):644–53. Available from: http://dx.doi.org/10.1080/15376516.2019.1646370 144. Xuguang H, Aofei T, Tao L, Longyan Z, Weijian B, Jiao G. Hesperidin ameliorates insulin resistance by regulating the IRS1-GLUT2 pathway via TLR4 in HepG2 cells. Phyther Res. 2019;33(6):1697–705. 145. Boro M, Balaji KN. CXCL1 and CXCL2 Regulate NLRP3 Inflammasome Activation via G-Protein–Coupled Receptor CXCR2. J Immunol. 2017;199(5):1660–71. 146. Guo LY, Yang F, Peng LJ, Li YB, Wang AP. CXCL2, a new critical factor and therapeutic target for cardiovascular diseases. Clin Exp Hypertens [Internet]. 2019;00(00):1–10. Available from: https://doi.org/10.1080/10641963.2019.1693585 147. de Jager SCA, Bongaerts BWC, Weber M, Kraaijeveld AO, Rousch M, Dimmeler S, et al. Chemokines CCL3/MIP1α, CCL5/RANTES and CCL18/PARC are Independent Risk Predictors of Short-Term Mortality in Patients with Acute Coronary Syndromes. PLoS One. 2012;7(9).

348

148. de Jager SCA, Kraaijeveld AO, Grauss RW, de Jager W, Liem SS, van der Hoeven BL, et al. CCL3 (MIP-1α) levels are elevated during acute coronary syndromes and show strong prognostic power for future ischemic events. J Mol Cell Cardiol. 2008;45(3):446–52. 149. Segiet A, Smykiewicz P, Kwiatkowski P, Żera T. Tumour necrosis factor and interleukin 10 in blood pressure regulation in spontaneously hypertensive and normotensive rats. Cytokine. 2019;113(April):185–94. 150. Mehaffey E, Majid DSA. Tumor necrosis factor-α, kidney function, and hypertension. Am J Physiol - Ren Physiol. 2017;313(4):F1005–8. 151. Libby P. Interleukin-1 Beta as a Target for Atherosclerosis Therapy. J Am Coll Cardiol. 2017;70(18):2278–89. 152. Qi J, Zhao XF, Yu XJ, Yi QY, Shi XL, Tan H, et al. Targeting Interleukin-1 beta to Suppress Sympathoexcitation in Hypothalamic Paraventricular Nucleus in Dahl Salt-Sensitive Hypertensive Rats. Cardiovasc Toxicol. 2016;16(3):298–306. 153. Desai SJ, Prickril B, Rasooly A. Mechanisms of phytonutrient modulation of Cyclooxygenase-2 (COX-2) and inflammation related to cancer. Nutr Cancer. 2018;70(3):350–75. 154. Miao Z, Guo M, Zhou S, Sun X, Wang F, Lu H, et al. Smoking and drinking influence the advancing of ischemic stroke disease by targeting PTGS2 and TNFAIP3. Exp Ther Med. 2018;16(1):61–6. 155. Ma A, Malynn BA. A20: Linking a complex regulator of ubiquitylation to immunity and human disease. Nat Rev

349

Immunol [Internet]. 2012;12(11):774–85. Available from: http://dx.doi.org/10.1038/nri3313 156. Khan TA, Sievenpiper JL. Controversies about sugars: results from systematic reviews and meta-analyses on obesity, cardiometabolic disease and diabetes. Eur J Nutr. 2016;55(s2):25–43. 157. Kühn EC, Slagman A, Kühn-Heid ECD, Seelig J, Schwiebert C, Minich WB, et al. Circulating levels of selenium-binding protein 1 (SELENBP1) are associated with risk for major adverse cardiac events and death. J Trace Elem Med Biol [Internet]. 2019;52(October 2018):247–53. Available from: https://doi.org/10.1016/j.jtemb.2019.01.005 158. Sun X, Haider Ali MSS, Moran M. The role of interactions of long non-coding RNAs and heterogeneous nuclear ribonucleoproteins in regulating cellular functions. Biochem J. 2017;474(17):2925–35. 159. Mathy NW, Chen XM. Long non-coding RNAs (lncRNAs) and their transcriptional control of inflammatory responses. J Biol Chem. 2017;292(30):12375–82. 160. Jiang C, Ding N, Li J, Jin X, Li L, Pan T, et al. Landscape of the long non-coding RNA transcriptome in human heart. Brief Bioinform. 2019;20(5):1812–25. 161. Esser D, Geleijnse JM, Matualatupauw JC, Dower JI, Kromhout D, Hollman PCH, et al. Pure flavonoid epicatechin and whole genome gene expression profiles in circulating immune cells in adults with elevated blood pressure: A randomised double-blind, placebo-controlled, crossover trial. PLoS One. 2018;13(4):1–15.

350

162. Osakabe N, Terao J. Possible mechanisms of postprandial physiological alterations following flavan 3-ol ingestion. Nutr Rev. 2018;76(3):174–86. 163. Wishart DS. Metabolomics for investigating physiological and pathophysiological processes. Physiol Rev. 2019;99(4):1819– 75. 164. Van QN, Veenstra TD, Issaq HJ. Metabolic profiling for the detection of bladder cancer. Curr Urol Rep. 2011;12(1):34–40. 165. Hanhineva K, Lankinen MA, Pedret A, Schwab U, Kolehmainen M, Paananen J, et al. Nontargeted metabolite profiling discriminates diet-specific biomarkers for consumption of whole grains, fatty fish, and bilberries in a randomized controlled trial. J Nutr. 2015;145(1):7–17. 166. Pla-Pagà L, Guirro M, Gual-Grau A, Gibert-Ramos A, Foguet- Romero E, Catalán Ú, et al. Proteomic Analysis of Heart and Kidney Tissues in Healthy and Metabolic Syndrome Rats after Hesperidin Supplementation. Mol Nutr Food Res. 2020;64(10):1–12. 167. Guirro M, Gual-grau A, Gibert-ramos A, Alcaide-hidalgo JM, Canela N, Mayneris-perxachs J. Metabolomics Elucidates Dose-Dependent Molecular Beneficial E ff ects of Hesperidin Supplementation in Rats Fed an Obesogenic Diet. (Cvd):1–19. 168. Manach C, Morand C, Gil-Izquierdo A, Bouteloup-Demange C, Rémésy C. Bioavailability in humans of the flavanones hesperidin and narirutin after the ingestion of two dose of orange juice. Eur J Clin Nutr. 2003;57(2):235–42. 169. Vallejo F, Larrosa M, Escudero E, Zafrilla MP, Cerdá B, Boza J, et al. Concentration and solubility of flavanones in orange

351

beverages affect their bioavailability in humans. J Agric Food Chem. 2010;58(10):6516–24. 170. Tomás-Navarro M, Vallejo F, Sentandreu E, Navarro JL, Tomás-Barberán FA. Volunteer stratification is more relevant than technological treatment in orange juice flavanone bioavailability. J Agric Food Chem. 2014;62(1):24–7. 171. Lloyd AJ, Beckmann M, Favé G, Mathers JC, Draper J. Proline betaine and its biotransformation products in fasting urine samples are potential biomarkers of habitual citrus fruit consumption. Br J Nutr. 2011;106(6):812–24. 172. Yang R, Dong J, Zhao H, Li H, Guo H, Wang S, et al. Association of branched-chain amino acids with carotid intima-media thickness and coronary artery disease risk factors. PLoS One. 2014;9(6). 173. Mirmiran P, Teymoori F, Asghari G, Azizi F. Dietary intakes of branched chain amino acids and the incidence of hypertension: A population-based prospective cohort study. Arch Iran Med. 2019;22(4):182–8. 174. Shannon M. Harlan, Deng-Fu Guo, Donald A. Morgan CF-S, Rahmouni1 and K. Hypothalamic mTORC1 Signaling Controls Sympathetic Nerve Activity and Arterial Pressure and Mediates Leptin Effects. Cell Metab. 2013;17(4):599–606. 175. B NC, An J, Bain JR, Muehlbauer MJ, Stevens RD, Lien LF, et al. A Branched-Chain Amino Acid-Related Metabolic Signature that Differentiates Obese and Lean Humans and Contributes to Insulin Resistance. Cell Metab. 9(4):311–26. 176. Connelly MA, Otvos JD, Shalaurova I, Playford MP, Mehta NN. GlycA, a novel biomarker of systemic inflammation and

352

cardiovascular disease risk. J Transl Med. 2017;15(1):1–5. 177. Topp H, Fusch G, Schöch G, Fusch C. Noninvasive markers of oxidative DNA stress, RNA degradation and protein degradation are differentially correlated with resting metabolic rate and energy intake in children and adolescents. Pediatr Res. 2008;64(3):246–50. 178. Montenegro Álvarez de Tejera P, Chamorro Merino G, Cabanes Mariscal M, Sánchez López P, Medina Font J. Development and validation of a method for simultaneous determination of pseudouridine and creatinine in human urine: Evaluation’s Index Pseudouridona/Creatinine in smokers and no-smokers. Sanid Mil. 2012;68(1):22–6. 179. Strifler G, Tuboly E, Görbe A, Boros M, Pécz D, Hartmann P. Targeting mitochondrial dysfunction with L-alpha glycerylphosphorylcholine. PLoS One. 2016;11(11):1–14. 180. Kymberleigh A. Romano, Ana Martinez-del Campo, Kazuyuki Kasahara, Carina L. Chittim, Eugenio I. Vivas, Daniel Amador-Noguez, Emily P. Balskus and FER. Metabolic, Epigenetic, and Transgenerational Effects of Gut Bacterial Choline Consumption. Cell Host Microbe. 2017;22(3):279–90. 181. Mair RD, Sirich TL, Meyer TW. Uremic toxin clearance and cardiovascular toxicities. Toxins (Basel). 2018;10(6):1–16. 182. Duranton F, Cohen G, De Smet R, Rodriguez M, Jankowski J, Vanholder R, et al. Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol. 2012;23(7):1258–70. 183. Hatem-Vaquero M, de Frutos S, Luengo A, González Abajo A, Griera M, Rodríguez-Puyol M, et al. Contribución de las toxinas urémicas a la fibrosis vascular asociada a la

353

enfermedad renal crónica. Nefrología. 2018;38(6):639–46. 184. Menni C, Mangino M, Cecelja M, Psatha M, Brosnan MJ, Trimmer J, et al. Metabolomic study of carotid-femoral pulse- wave velocity in women. J Hypertens. 2015;33(4):791–6. 185. Ottosson F, Brunkwall L, Smith E, Orho-Melander M, Nilsson PM, Fernandez C, et al. The gut microbiota-related metabolite phenylacetylglutamine associates with increased risk of incident coronary artery disease. J Hypertens. 2020;Publish Ah. 186. Wu IW, Hsu KH, Lee CC, Sun CY, Hsu HJ, Tsai CJ, et al. P- cresyl sulphate and indoxyl sulphate predict progression of chronic kidney disease. Nephrol Dial Transplant. 2011;26(3):938–47. 187. Gormsen LC, Svart M, Thomsen HH, Søndergaard E, Vendelbo MH, Christensen N, et al. Ketone body infusion with 3-hydroxybutyrate reduces myocardial glucose uptake and increases blood flow in humans: A positron emission tomography study. J Am Heart Assoc. 2017;6(3):1–11. 188. Li M, Wang B, Zhang M, Rantalainen M, Wang S, Zhou H, et al. Symbiotic gut microbes modulate human metabolic phenotypes. Proc Natl Acad Sci U S A. 2008;105(6):2117–22. 189. Vangipurapu J, Silva LF, Kuulasmaa T, Smith U, Laakso M. Microbiota-related metabolites and the risk of type 2 diabetes. Diabetes Care. 2020;43(6):1319–25. 190. Yari Z, Cheraghpour M, Hekmatdoost A. Flaxseed and/or hesperidin supplementation in metabolic syndrome: an open- labeled randomized controlled trial. Eur J Nutr [Internet]. 2020;(0123456789). Available from:

354

https://doi.org/10.1007/s00394-020-02246-9 191. Yari Z, Movahedian M, Imani H, Alavian SM, Hedayati M, Hekmatdoost A. The effect of hesperidin supplementation on metabolic profiles in patients with metabolic syndrome: a randomized, double-blind, placebo-controlled clinical trial. Eur J Nutr [Internet]. 2020;59(6):2569–77. Available from: https://doi.org/10.1007/s00394-019-02105-2 192. Takumi H, Nakamura H, Simizu T, Harada R, Kometani T, Nadamoto T, et al. Bioavailability of orally administered water-dispersible hesperetin and its effect on peripheral vasodilatation in human subjects: Implication of endothelial functions of plasma conjugated metabolites. Food Funct. 2012;3(4):389–98. 193. Dobiaš L, Petrová M, Vojtko R, Kristová V. Long-term Treatment with Hesperidin Improves Endothelium-dependent Vasodilation in Femoral Artery of Spontaneously Hypertensive

Rats: The Involvement of NO-synthase and K v Channels. Phyther Res [Internet]. 2016;30(10):1665–71. Available from: http://www.ncbi.nlm.nih.gov/pubmed/27363952%0Ahttp://doi. wiley.com/10.1002/ptr.5670 194. Parlak Dr. A, Aydoǧan Ü, Iyisoy A, Dikililer MA, Kut A, Çakir E, et al. Elevated pentraxin-3 levels are related to blood pressure levels in hypertensive patients: An observational study. Anadolu Kardiyol Derg. 2012;12(4):298–304. 195. Gunes F, Akbal E, Cakir E, Akyurek O. Visfatin May Be a Novel Marker for Identifying Stages of Essential Hypertension in Advanced Age Patients. 2011;553–7. 196. Allison M, Blumenthal RS, Cannon CP, Lemos JA De,

355

Findeiss L, Gore JM, et al. Treatment of Hypertension in Patients With Coronary Artery Disease. 2015;65(18). 197. Jamshidi N, Mantri N, Cohen MM. Acute effects of dietary plant nutrients on transcriptome profiles: evidence from human studies. Crit Rev Food Sci Nutr [Internet]. 2020;60(11):1869– 80. Available from: https://doi.org/10.1080/10408398.2019.1608154 198. Reagan-Shaw S, Nihal M, Ahmad N. Dose translation from animal to human studies revisited. FASEB J. 2008;22(3):659– 61. 199. Brisdelli F, Francesco L Di, Giorgi A, Lizzi AR, Luzi C, Mignogna G, et al. Proteomic Analysis of Quercetin-Treated K562 Cells. 2019; 200. Ali F, Dong S, Park J, Ok P. Identification of Proteins Differentially Expressed by Quercetin Treatment in a Middle Cerebral Artery Occlusion Model : A Proteomics Approach. Neurochem Res [Internet]. 2018;0(0):0. Available from: http://dx.doi.org/10.1007/s11064-018-2576-x 201. Catherine A. Lozupone, Jesse I. Stombaugh, Jeffrey I. Gordon, Janet K. Jansson A, Knight R. Diversity, stability and resilience of the human gut microbiota. Nature. 2012;489(7415):220–30. 202. Vargova V, Pytliak M, Mechirova V, Cao W, Hu SS, Ye LH, et al. Bioavailability in humans of the flavanones hesperidin and narirutin after the ingestion of two doses of orange juice. J Agric Food Chem [Internet]. 2016;4(3):228–33. Available from: http://dx.doi.org/10.1016/j.fct.2011.11.038%255Cnhttp://dx.do

356

i.org/10.1016/j.lfs.2014.12.030%255Cnhttp://pubs.acs.org/doi/ abs/10.1021/acssuschemeng.6b00626%255Cnhttp://pubs.acs.o rg/doi/abs/10.1021/acs.jafc.6b02183%255Cnhttp://www.ncbi.n lm.nih.gov/pubmed/20716914%25 203. Patra AK. Dietary phytochemicals and microbes. Vol. 9789400739, Dietary Phytochemicals and Microbes. 2012. 1– 400 p. 204. Guirro M, Torrell H, Costa A, Mayneris-Perxachs J, Herrero P, Gual-Grau A, et al. Multi-omics approach to elucidate the gut microbiota activity: Metaproteomics and metagenomics connection. Electrophoresis. 2018;39(13):1692–701. 205. Leon-Mimila P, Wang J, Huertas-Vazquez A. Relevance of Multi-Omics Studies in Cardiovascular Diseases. Front Cardiovasc Med. 2019;6(July). 206. Ferreira de Oliveira JMP, Santos C, Fernandes E. Therapeutic potential of hesperidin and its aglycone hesperetin: Cell cycle regulation and apoptosis induction in cancer models. Phytomedicine [Internet]. 2019;152887. Available from: https://doi.org/10.1016/j.phymed.2019.152887 207. Aggarwal V, Tuli HS, Thakral F, Singhal P, Aggarwal D, Srivastava S, et al. Molecular mechanisms of action of hesperidin in cancer: Recent trends and advancements. Exp Biol Med. 2020;245(5):486–97. 208. Balmeh N, Mahmoudi S, Mohammadi N, Karabedianhajiabadi A. Predicted therapeutic targets for COVID-19 disease by inhibiting SARS-CoV-2 and its related receptors. 2020;(January). 209. Bellavite P, Donzelli A. Hesperidin and SARS-CoV-2: New

357

light on the healthy function of citrus fruits. Antioxidants. 2020;9(8):1–18. 210. Yusuf A. Haggaga, Nahla E. El-Ashmawyb KMO. Is hesperidin essential for prophylaxis and treatment of COVID- 19 Infection? 2020;144(January).

358

359

ANNEX

360

361

VISIT TO OTHER INTERNATIONAL CENTRE:

Visit to the International Agency for Research on Cancer / World Health Organization (IARC/WHO) of Lyon, France. Period: October 8th, 2018 – February 21st, 2019.

362